[3/4] xfs: support bulk loading of staged btrees

Commit Message

Darrick J. Wong Oct. 9, 2019, 4:48 p.m. UTC

From: Darrick J. Wong <[e-mail]>

Add a new btree function that enables us to bulk load a btree cursor.
This will be used by the upcoming online repair patches to generate new
btrees.

Signed-off-by: Darrick J. Wong <[e-mail]>
---
 fs/xfs/libxfs/xfs_btree.c |  566 +++++++++++++++++++++++++++++++++++++++++++++
 fs/xfs/libxfs/xfs_btree.h |   43 +++
 fs/xfs/xfs_trace.c        |    1 
 fs/xfs/xfs_trace.h        |   85 +++++++
 4 files changed, 694 insertions(+), 1 deletion(-)

Comments

Brian Foster Oct. 16, 2019, 3:26 p.m. UTC | #1

On Wed, Oct 09, 2019 at 09:48:18AM -0700, Darrick J. Wong wrote:
> From: Darrick J. Wong <[e-mail]>
> 
> Add a new btree function that enables us to bulk load a btree cursor.
> This will be used by the upcoming online repair patches to generate new
> btrees.
> 
> Signed-off-by: Darrick J. Wong <[e-mail]>
> ---
>  fs/xfs/libxfs/xfs_btree.c |  566 +++++++++++++++++++++++++++++++++++++++++++++
>  fs/xfs/libxfs/xfs_btree.h |   43 +++
>  fs/xfs/xfs_trace.c        |    1 
>  fs/xfs/xfs_trace.h        |   85 +++++++
>  4 files changed, 694 insertions(+), 1 deletion(-)
> 
> 
> diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
> index 4b06d5d86834..17b0fdb87729 100644
> --- a/fs/xfs/libxfs/xfs_btree.c
> +++ b/fs/xfs/libxfs/xfs_btree.c
...
> @@ -5104,3 +5104,567 @@ xfs_btree_commit_ifakeroot(
>  	cur->bc_ops = ops;
>  	cur->bc_flags &= ~XFS_BTREE_STAGING;
>  }
> +
> +/*
> + * Bulk Loading of Staged Btrees
> + * =============================
> + *
> + * This interface is used with a staged btree cursor to create a totally new
> + * btree with a large number of records (i.e. more than what would fit in a
> + * single block).  When the creation is complete, the new root can be linked
> + * atomically into the filesystem by committing the staged cursor.
> + *

Thanks for the documentation. So what is the purpose behind the whole
bulk loading thing as opposed to something like faking up an AG
structure (i.e. AGF) somewhere and using the existing cursor mechanisms
(or something closer to it) to copy records from one place to another?
Is it purely a performance/efficiency tradeoff? Bulk block allocation
issues? Transactional/atomicity issues? All (or none :P) of the above?

This is my first pass through this so I'm mostly looking at big picture
until I get to a point to see how these bits are used. The mechanism
itself seems reasonable in principle, but the reason I ask is it also
seems like there's inherent value in using more of same infrastructure
to reconstruct a tree that we use to create one in the first place. We
also already have primitives for things like fork swapping via the
extent swap mechanism, etc.

Brian

> + * The first step for the caller is to construct a fake btree root structure
> + * and a staged btree cursor.  A staging cursor contains all the geometry
> + * information for the btree type but will fail all operations that could have
> + * side effects in the filesystem (e.g. btree shape changes).  Regular
> + * operations will not work unless the staging cursor is committed and becomes
> + * a regular cursor.
> + *
> + * For a btree rooted in an AG header, use an xbtree_afakeroot structure.
> + * This should be initialized to zero.  For a btree rooted in an inode fork,
> + * use an xbtree_ifakeroot structure.  @if_fork_size field should be set to
> + * the number of bytes available to the fork in the inode; @if_fork should
> + * point to a freshly allocated xfs_inode_fork; and @if_format should be set
> + * to the appropriate fork type (e.g. XFS_DINODE_FMT_BTREE).
> + *
> + * The next step for the caller is to initialize a struct xfs_btree_bload
> + * context.  The @nr_records field is the number of records that are to be
> + * loaded into the btree.  The @leaf_slack and @node_slack fields are the
> + * number of records (or key/ptr) slots to leave empty in new btree blocks.
> + * If a caller sets a slack value to -1, the slack value will be computed to
> + * fill the block halfway between minrecs and maxrecs items per block.
> + *
> + * The number of items placed in each btree block is computed via the following
> + * algorithm: For leaf levels, the number of items for the level is nr_records.
> + * For node levels, the number of items for the level is the number of blocks
> + * in the next lower level of the tree.  For each level, the desired number of
> + * items per block is defined as:
> + *
> + * desired = max(minrecs, maxrecs - slack factor)
> + *
> + * The number of blocks for the level is defined to be:
> + *
> + * blocks = nr_items / desired
> + *
> + * Note this is rounded down so that the npb calculation below will never fall
> + * below minrecs.  The number of items that will actually be loaded into each
> + * btree block is defined as:
> + *
> + * npb =  nr_items / blocks
> + *
> + * Some of the leftmost blocks in the level will contain one extra record as
> + * needed to handle uneven division.  If the number of records in any block
> + * would exceed maxrecs for that level, blocks is incremented and npb is
> + * recalculated.
> + *
> + * In other words, we compute the number of blocks needed to satisfy a given
> + * loading level, then spread the items as evenly as possible.
> + *
> + * To complete this step, call xfs_btree_bload_compute_geometry, which uses
> + * those settings to compute the height of the btree and the number of blocks
> + * that will be needed to construct the btree.  These values are stored in the
> + * @btree_height and @nr_blocks fields.
> + *
> + * At this point, the caller must allocate @nr_blocks blocks and save them for
> + * later.  If space is to be allocated transactionally, the staging cursor
> + * must be deleted before and recreated after, which is why computing the
> + * geometry is a separate step.
> + *
> + * The fourth step in the bulk loading process is to set the function pointers
> + * in the bload context structure.  @get_data will be called for each record
> + * that will be loaded into the btree; it should set the cursor's bc_rec
> + * field, which will be converted to on-disk format and copied into the
> + * appropriate record slot.  @alloc_block should supply one of the blocks
> + * allocated in the previous step.  For btrees which are rooted in an inode
> + * fork, @iroot_size is called to compute the size of the incore btree root
> + * block.  Call xfs_btree_bload to start constructing the btree.
> + *
> + * The final step is to commit the staging cursor, which logs the new btree
> + * root and turns the btree into a regular btree cursor, and free the fake
> + * roots.
> + */
> +
> +/*
> + * Put a btree block that we're loading onto the ordered list and release it.
> + * The btree blocks will be written when the final transaction swapping the
> + * btree roots is committed.
> + */
> +static void
> +xfs_btree_bload_drop_buf(
> +	struct xfs_trans	*tp,
> +	struct xfs_buf		**bpp)
> +{
> +	if (*bpp == NULL)
> +		return;
> +
> +	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_BTREE_BUF);
> +	xfs_trans_ordered_buf(tp, *bpp);
> +	xfs_trans_brelse(tp, *bpp);
> +	*bpp = NULL;
> +}
> +
> +/* Allocate and initialize one btree block for bulk loading. */
> +STATIC int
> +xfs_btree_bload_prep_block(
> +	struct xfs_btree_cur		*cur,
> +	struct xfs_btree_bload		*bbl,
> +	unsigned int			level,
> +	unsigned int			nr_this_block,
> +	union xfs_btree_ptr		*ptrp,
> +	struct xfs_buf			**bpp,
> +	struct xfs_btree_block		**blockp,
> +	void				*priv)
> +{
> +	union xfs_btree_ptr		new_ptr;
> +	struct xfs_buf			*new_bp;
> +	struct xfs_btree_block		*new_block;
> +	int				ret;
> +
> +	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
> +	    level == cur->bc_nlevels - 1) {
> +		struct xfs_ifork	*ifp = cur->bc_private.b.ifake->if_fork;
> +		size_t			new_size;
> +
> +		/* Allocate a new incore btree root block. */
> +		new_size = bbl->iroot_size(cur, nr_this_block, priv);
> +		ifp->if_broot = kmem_zalloc(new_size, 0);
> +		ifp->if_broot_bytes = (int)new_size;
> +		ifp->if_flags |= XFS_IFBROOT;
> +
> +		/* Initialize it and send it out. */
> +		xfs_btree_init_block_int(cur->bc_mp, ifp->if_broot,
> +				XFS_BUF_DADDR_NULL, cur->bc_btnum, level,
> +				nr_this_block, cur->bc_private.b.ip->i_ino,
> +				cur->bc_flags);
> +
> +		*bpp = NULL;
> +		*blockp = ifp->if_broot;
> +		xfs_btree_set_ptr_null(cur, ptrp);
> +		return 0;
> +	}
> +
> +	/* Allocate a new leaf block. */
> +	ret = bbl->alloc_block(cur, &new_ptr, priv);
> +	if (ret)
> +		return ret;
> +
> +	ASSERT(!xfs_btree_ptr_is_null(cur, &new_ptr));
> +
> +	ret = xfs_btree_get_buf_block(cur, &new_ptr, &new_block, &new_bp);
> +	if (ret)
> +		return ret;
> +
> +	/* Initialize the btree block. */
> +	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
> +	if (*blockp)
> +		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
> +	xfs_btree_set_sibling(cur, new_block, ptrp, XFS_BB_LEFTSIB);
> +	xfs_btree_set_numrecs(new_block, nr_this_block);
> +
> +	/* Release the old block and set the out parameters. */
> +	xfs_btree_bload_drop_buf(cur->bc_tp, bpp);
> +	*blockp = new_block;
> +	*bpp = new_bp;
> +	xfs_btree_copy_ptrs(cur, ptrp, &new_ptr, 1);
> +	return 0;
> +}
> +
> +/* Load one leaf block. */
> +STATIC int
> +xfs_btree_bload_leaf(
> +	struct xfs_btree_cur		*cur,
> +	unsigned int			recs_this_block,
> +	xfs_btree_bload_get_fn		get_data,
> +	struct xfs_btree_block		*block,
> +	void				*priv)
> +{
> +	unsigned int			j;
> +	int				ret;
> +
> +	/* Fill the leaf block with records. */
> +	for (j = 1; j <= recs_this_block; j++) {
> +		union xfs_btree_rec	*block_recs;
> +
> +		ret = get_data(cur, priv);
> +		if (ret)
> +			return ret;
> +		block_recs = xfs_btree_rec_addr(cur, j, block);
> +		cur->bc_ops->init_rec_from_cur(cur, block_recs);
> +	}
> +
> +	return 0;
> +}
> +
> +/* Load one node block. */
> +STATIC int
> +xfs_btree_bload_node(
> +	struct xfs_btree_cur	*cur,
> +	unsigned int		recs_this_block,
> +	union xfs_btree_ptr	*child_ptr,
> +	struct xfs_btree_block	*block)
> +{
> +	unsigned int		j;
> +	int			ret;
> +
> +	/* Fill the node block with keys and pointers. */
> +	for (j = 1; j <= recs_this_block; j++) {
> +		union xfs_btree_key	child_key;
> +		union xfs_btree_ptr	*block_ptr;
> +		union xfs_btree_key	*block_key;
> +		struct xfs_btree_block	*child_block;
> +		struct xfs_buf		*child_bp;
> +
> +		ASSERT(!xfs_btree_ptr_is_null(cur, child_ptr));
> +
> +		ret = xfs_btree_get_buf_block(cur, child_ptr, &child_block,
> +				&child_bp);
> +		if (ret)
> +			return ret;
> +
> +		xfs_btree_get_keys(cur, child_block, &child_key);
> +
> +		block_ptr = xfs_btree_ptr_addr(cur, j, block);
> +		xfs_btree_copy_ptrs(cur, block_ptr, child_ptr, 1);
> +
> +		block_key = xfs_btree_key_addr(cur, j, block);
> +		xfs_btree_copy_keys(cur, block_key, &child_key, 1);
> +
> +		xfs_btree_get_sibling(cur, child_block, child_ptr,
> +				XFS_BB_RIGHTSIB);
> +		xfs_trans_brelse(cur->bc_tp, child_bp);
> +	}
> +
> +	return 0;
> +}
> +
> +/*
> + * Compute the maximum number of records (or keyptrs) per block that we want to
> + * install at this level in the btree.  Caller is responsible for having set
> + * @cur->bc_private.b.forksize to the desired fork size, if appropriate.
> + */
> +STATIC unsigned int
> +xfs_btree_bload_max_npb(
> +	struct xfs_btree_cur	*cur,
> +	struct xfs_btree_bload	*bbl,
> +	unsigned int		level)
> +{
> +	unsigned int		ret;
> +
> +	if (level == cur->bc_nlevels - 1 && cur->bc_ops->get_dmaxrecs)
> +		return cur->bc_ops->get_dmaxrecs(cur, level);
> +
> +	ret = cur->bc_ops->get_maxrecs(cur, level);
> +	if (level == 0)
> +		ret -= bbl->leaf_slack;
> +	else
> +		ret -= bbl->node_slack;
> +	return ret;
> +}
> +
> +/*
> + * Compute the desired number of records (or keyptrs) per block that we want to
> + * install at this level in the btree, which must be somewhere between minrecs
> + * and max_npb.  The caller is free to install fewer records per block.
> + */
> +STATIC unsigned int
> +xfs_btree_bload_desired_npb(
> +	struct xfs_btree_cur	*cur,
> +	struct xfs_btree_bload	*bbl,
> +	unsigned int		level)
> +{
> +	unsigned int		npb = xfs_btree_bload_max_npb(cur, bbl, level);
> +
> +	/* Root blocks are not subject to minrecs rules. */
> +	if (level == cur->bc_nlevels - 1)
> +		return max(1U, npb);
> +
> +	return max_t(unsigned int, cur->bc_ops->get_minrecs(cur, level), npb);
> +}
> +
> +/*
> + * Compute the number of records to be stored in each block at this level and
> + * the number of blocks for this level.  For leaf levels, we must populate an
> + * empty root block even if there are no records, so we have to have at least
> + * one block.
> + */
> +STATIC void
> +xfs_btree_bload_level_geometry(
> +	struct xfs_btree_cur	*cur,
> +	struct xfs_btree_bload	*bbl,
> +	unsigned int		level,
> +	uint64_t		nr_this_level,
> +	unsigned int		*avg_per_block,
> +	uint64_t		*blocks,
> +	uint64_t		*blocks_with_extra)
> +{
> +	uint64_t		npb;
> +	uint64_t		dontcare;
> +	unsigned int		desired_npb;
> +	unsigned int		maxnr;
> +
> +	maxnr = cur->bc_ops->get_maxrecs(cur, level);
> +
> +	/*
> +	 * Compute the number of blocks we need to fill each block with the
> +	 * desired number of records/keyptrs per block.  Because desired_npb
> +	 * could be minrecs, we use regular integer division (which rounds
> +	 * the block count down) so that in the next step the effective # of
> +	 * items per block will never be less than desired_npb.
> +	 */
> +	desired_npb = xfs_btree_bload_desired_npb(cur, bbl, level);
> +	*blocks = div64_u64_rem(nr_this_level, desired_npb, &dontcare);
> +	*blocks = max(1ULL, *blocks);
> +
> +	/*
> +	 * Compute the number of records that we will actually put in each
> +	 * block, assuming that we want to spread the records evenly between
> +	 * the blocks.  Take care that the effective # of items per block (npb)
> +	 * won't exceed maxrecs even for the blocks that get an extra record,
> +	 * since desired_npb could be maxrecs, and in the previous step we
> +	 * rounded the block count down.
> +	 */
> +	npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> +	if (npb > maxnr || (npb == maxnr && *blocks_with_extra > 0)) {
> +		(*blocks)++;
> +		npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> +	}
> +
> +	*avg_per_block = min_t(uint64_t, npb, nr_this_level);
> +
> +	trace_xfs_btree_bload_level_geometry(cur, level, nr_this_level,
> +			*avg_per_block, desired_npb, *blocks,
> +			*blocks_with_extra);
> +}
> +
> +/*
> + * Ensure a slack value is appropriate for the btree.
> + *
> + * If the slack value is negative, set slack so that we fill the block to
> + * halfway between minrecs and maxrecs.  Make sure the slack is never so large
> + * that we can underflow minrecs.
> + */
> +static void
> +xfs_btree_bload_ensure_slack(
> +	struct xfs_btree_cur	*cur,
> +	int			*slack,
> +	int			level)
> +{
> +	int			maxr;
> +	int			minr;
> +
> +	/*
> +	 * We only care about slack for btree blocks, so set the btree nlevels
> +	 * to 3 so that level 0 is a leaf block and level 1 is a node block.
> +	 * Avoid straying into inode roots, since we don't do slack there.
> +	 */
> +	cur->bc_nlevels = 3;
> +	maxr = cur->bc_ops->get_maxrecs(cur, level);
> +	minr = cur->bc_ops->get_minrecs(cur, level);
> +
> +	/*
> +	 * If slack is negative, automatically set slack so that we load the
> +	 * btree block approximately halfway between minrecs and maxrecs.
> +	 * Generally, this will net us 75% loading.
> +	 */
> +	if (*slack < 0)
> +		*slack = maxr - ((maxr + minr) >> 1);
> +
> +	*slack = min(*slack, maxr - minr);
> +}
> +
> +/*
> + * Prepare a btree cursor for a bulk load operation by computing the geometry
> + * fields in @bbl.  Caller must ensure that the btree cursor is a staging
> + * cursor.  This function can be called multiple times.
> + */
> +int
> +xfs_btree_bload_compute_geometry(
> +	struct xfs_btree_cur	*cur,
> +	struct xfs_btree_bload	*bbl,
> +	uint64_t		nr_records)
> +{
> +	uint64_t		nr_blocks = 0;
> +	uint64_t		nr_this_level;
> +
> +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> +
> +	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
> +	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
> +
> +	bbl->nr_records = nr_this_level = nr_records;
> +	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
> +		uint64_t	level_blocks;
> +		uint64_t	dontcare64;
> +		unsigned int	level = cur->bc_nlevels - 1;
> +		unsigned int	avg_per_block;
> +
> +		/*
> +		 * If all the things we want to store at this level would fit
> +		 * in a single root block, then we have our btree root and are
> +		 * done.  Note that bmap btrees do not allow records in the
> +		 * root.
> +		 */
> +		if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) || level != 0) {
> +			xfs_btree_bload_level_geometry(cur, bbl, level,
> +					nr_this_level, &avg_per_block,
> +					&level_blocks, &dontcare64);
> +			if (nr_this_level <= avg_per_block) {
> +				nr_blocks++;
> +				break;
> +			}
> +		}
> +
> +		/*
> +		 * Otherwise, we have to store all the records for this level
> +		 * in blocks and therefore need another level of btree to point
> +		 * to those blocks.  Increase the number of levels and
> +		 * recompute the number of records we can store at this level
> +		 * because that can change depending on whether or not a level
> +		 * is the root level.
> +		 */
> +		cur->bc_nlevels++;
> +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> +				&avg_per_block, &level_blocks, &dontcare64);
> +		nr_blocks += level_blocks;
> +		nr_this_level = level_blocks;
> +	}
> +
> +	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
> +		return -EOVERFLOW;
> +
> +	bbl->btree_height = cur->bc_nlevels;
> +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
> +		bbl->nr_blocks = nr_blocks - 1;
> +	else
> +		bbl->nr_blocks = nr_blocks;
> +	return 0;
> +}
> +
> +/*
> + * Bulk load a btree.
> + *
> + * Load @bbl->nr_records quantity of records into a btree using the supplied
> + * empty and staging btree cursor @cur and a @bbl that has been filled out by
> + * the xfs_btree_bload_compute_geometry function.
> + *
> + * The @bbl->get_data function must populate the cursor's bc_rec every time it
> + * is called.  The @bbl->alloc_block function will be used to allocate new
> + * btree blocks.  @priv is passed to both functions.
> + *
> + * Caller must ensure that @cur is a staging cursor.  Any existing btree rooted
> + * in the fakeroot will be lost, so do not call this function twice.
> + */
> +int
> +xfs_btree_bload(
> +	struct xfs_btree_cur		*cur,
> +	struct xfs_btree_bload		*bbl,
> +	void				*priv)
> +{
> +	union xfs_btree_ptr		child_ptr;
> +	union xfs_btree_ptr		ptr;
> +	struct xfs_buf			*bp = NULL;
> +	struct xfs_btree_block		*block = NULL;
> +	uint64_t			nr_this_level = bbl->nr_records;
> +	uint64_t			blocks;
> +	uint64_t			i;
> +	uint64_t			blocks_with_extra;
> +	uint64_t			total_blocks = 0;
> +	unsigned int			avg_per_block;
> +	unsigned int			level = 0;
> +	int				ret;
> +
> +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> +
> +	cur->bc_nlevels = bbl->btree_height;
> +	xfs_btree_set_ptr_null(cur, &child_ptr);
> +	xfs_btree_set_ptr_null(cur, &ptr);
> +
> +	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> +			&avg_per_block, &blocks, &blocks_with_extra);
> +
> +	/* Load each leaf block. */
> +	for (i = 0; i < blocks; i++) {
> +		unsigned int		nr_this_block = avg_per_block;
> +
> +		if (i < blocks_with_extra)
> +			nr_this_block++;
> +
> +		ret = xfs_btree_bload_prep_block(cur, bbl, level,
> +				nr_this_block, &ptr, &bp, &block, priv);
> +		if (ret)
> +			return ret;
> +
> +		trace_xfs_btree_bload_block(cur, level, i, blocks, &ptr,
> +				nr_this_block);
> +
> +		ret = xfs_btree_bload_leaf(cur, nr_this_block, bbl->get_data,
> +				block, priv);
> +		if (ret)
> +			goto out;
> +
> +		/* Record the leftmost pointer to start the next level. */
> +		if (i == 0)
> +			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
> +	}
> +	total_blocks += blocks;
> +	xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> +
> +	/* Populate the internal btree nodes. */
> +	for (level = 1; level < cur->bc_nlevels; level++) {
> +		union xfs_btree_ptr	first_ptr;
> +
> +		nr_this_level = blocks;
> +		block = NULL;
> +		xfs_btree_set_ptr_null(cur, &ptr);
> +
> +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> +				&avg_per_block, &blocks, &blocks_with_extra);
> +
> +		/* Load each node block. */
> +		for (i = 0; i < blocks; i++) {
> +			unsigned int	nr_this_block = avg_per_block;
> +
> +			if (i < blocks_with_extra)
> +				nr_this_block++;
> +
> +			ret = xfs_btree_bload_prep_block(cur, bbl, level,
> +					nr_this_block, &ptr, &bp, &block,
> +					priv);
> +			if (ret)
> +				return ret;
> +
> +			trace_xfs_btree_bload_block(cur, level, i, blocks,
> +					&ptr, nr_this_block);
> +
> +			ret = xfs_btree_bload_node(cur, nr_this_block,
> +					&child_ptr, block);
> +			if (ret)
> +				goto out;
> +
> +			/*
> +			 * Record the leftmost pointer to start the next level.
> +			 */
> +			if (i == 0)
> +				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
> +		}
> +		total_blocks += blocks;
> +		xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> +		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
> +	}
> +
> +	/* Initialize the new root. */
> +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
> +		ASSERT(xfs_btree_ptr_is_null(cur, &ptr));
> +		cur->bc_private.b.ifake->if_levels = cur->bc_nlevels;
> +		cur->bc_private.b.ifake->if_blocks = total_blocks - 1;
> +	} else {
> +		cur->bc_private.a.afake->af_root = be32_to_cpu(ptr.s);
> +		cur->bc_private.a.afake->af_levels = cur->bc_nlevels;
> +		cur->bc_private.a.afake->af_blocks = total_blocks;
> +	}
> +out:
> +	if (bp)
> +		xfs_trans_brelse(cur->bc_tp, bp);
> +	return ret;
> +}
> diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
> index a17becb72ab8..5c6992a04ea2 100644
> --- a/fs/xfs/libxfs/xfs_btree.h
> +++ b/fs/xfs/libxfs/xfs_btree.h
> @@ -582,4 +582,47 @@ void xfs_btree_stage_ifakeroot(struct xfs_btree_cur *cur,
>  void xfs_btree_commit_ifakeroot(struct xfs_btree_cur *cur, int whichfork,
>  		const struct xfs_btree_ops *ops);
>  
> +typedef int (*xfs_btree_bload_get_fn)(struct xfs_btree_cur *cur, void *priv);
> +typedef int (*xfs_btree_bload_alloc_block_fn)(struct xfs_btree_cur *cur,
> +		union xfs_btree_ptr *ptr, void *priv);
> +typedef size_t (*xfs_btree_bload_iroot_size_fn)(struct xfs_btree_cur *cur,
> +		unsigned int nr_this_level, void *priv);
> +
> +/* Bulk loading of staged btrees. */
> +struct xfs_btree_bload {
> +	/* Function to store a record in the cursor. */
> +	xfs_btree_bload_get_fn		get_data;
> +
> +	/* Function to allocate a block for the btree. */
> +	xfs_btree_bload_alloc_block_fn	alloc_block;
> +
> +	/* Function to compute the size of the in-core btree root block. */
> +	xfs_btree_bload_iroot_size_fn	iroot_size;
> +
> +	/* Number of records the caller wants to store. */
> +	uint64_t			nr_records;
> +
> +	/* Number of btree blocks needed to store those records. */
> +	uint64_t			nr_blocks;
> +
> +	/*
> +	 * Number of free records to leave in each leaf block.  If this (or
> +	 * any of the slack values) are negative, this will be computed to
> +	 * be halfway between maxrecs and minrecs.  This typically leaves the
> +	 * block 75% full.
> +	 */
> +	int				leaf_slack;
> +
> +	/* Number of free keyptrs to leave in each node block. */
> +	int				node_slack;
> +
> +	/* Computed btree height. */
> +	unsigned int			btree_height;
> +};
> +
> +int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
> +		struct xfs_btree_bload *bbl, uint64_t nr_records);
> +int xfs_btree_bload(struct xfs_btree_cur *cur, struct xfs_btree_bload *bbl,
> +		void *priv);
> +
>  #endif	/* __XFS_BTREE_H__ */
> diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
> index bc85b89f88ca..9b5e58a92381 100644
> --- a/fs/xfs/xfs_trace.c
> +++ b/fs/xfs/xfs_trace.c
> @@ -6,6 +6,7 @@
>  #include "xfs.h"
>  #include "xfs_fs.h"
>  #include "xfs_shared.h"
> +#include "xfs_bit.h"
>  #include "xfs_format.h"
>  #include "xfs_log_format.h"
>  #include "xfs_trans_resv.h"
> diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
> index a78055521fcd..6d7ba64b7a0f 100644
> --- a/fs/xfs/xfs_trace.h
> +++ b/fs/xfs/xfs_trace.h
> @@ -35,6 +35,7 @@ struct xfs_icreate_log;
>  struct xfs_owner_info;
>  struct xfs_trans_res;
>  struct xfs_inobt_rec_incore;
> +union xfs_btree_ptr;
>  
>  DECLARE_EVENT_CLASS(xfs_attr_list_class,
>  	TP_PROTO(struct xfs_attr_list_context *ctx),
> @@ -3670,6 +3671,90 @@ TRACE_EVENT(xfs_btree_commit_ifakeroot,
>  		  __entry->blocks)
>  )
>  
> +TRACE_EVENT(xfs_btree_bload_level_geometry,
> +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> +		 uint64_t nr_this_level, unsigned int nr_per_block,
> +		 unsigned int desired_npb, uint64_t blocks,
> +		 uint64_t blocks_with_extra),
> +	TP_ARGS(cur, level, nr_this_level, nr_per_block, desired_npb, blocks,
> +		blocks_with_extra),
> +	TP_STRUCT__entry(
> +		__field(dev_t, dev)
> +		__field(xfs_btnum_t, btnum)
> +		__field(unsigned int, level)
> +		__field(unsigned int, nlevels)
> +		__field(uint64_t, nr_this_level)
> +		__field(unsigned int, nr_per_block)
> +		__field(unsigned int, desired_npb)
> +		__field(unsigned long long, blocks)
> +		__field(unsigned long long, blocks_with_extra)
> +	),
> +	TP_fast_assign(
> +		__entry->dev = cur->bc_mp->m_super->s_dev;
> +		__entry->btnum = cur->bc_btnum;
> +		__entry->level = level;
> +		__entry->nlevels = cur->bc_nlevels;
> +		__entry->nr_this_level = nr_this_level;
> +		__entry->nr_per_block = nr_per_block;
> +		__entry->desired_npb = desired_npb;
> +		__entry->blocks = blocks;
> +		__entry->blocks_with_extra = blocks_with_extra;
> +	),
> +	TP_printk("dev %d:%d btree %s level %u/%u nr_this_level %llu nr_per_block %u desired_npb %u blocks %llu blocks_with_extra %llu",
> +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> +		  __entry->level,
> +		  __entry->nlevels,
> +		  __entry->nr_this_level,
> +		  __entry->nr_per_block,
> +		  __entry->desired_npb,
> +		  __entry->blocks,
> +		  __entry->blocks_with_extra)
> +)
> +
> +TRACE_EVENT(xfs_btree_bload_block,
> +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> +		 uint64_t block_idx, uint64_t nr_blocks,
> +		 union xfs_btree_ptr *ptr, unsigned int nr_records),
> +	TP_ARGS(cur, level, block_idx, nr_blocks, ptr, nr_records),
> +	TP_STRUCT__entry(
> +		__field(dev_t, dev)
> +		__field(xfs_btnum_t, btnum)
> +		__field(unsigned int, level)
> +		__field(unsigned long long, block_idx)
> +		__field(unsigned long long, nr_blocks)
> +		__field(xfs_agnumber_t, agno)
> +		__field(xfs_agblock_t, agbno)
> +		__field(unsigned int, nr_records)
> +	),
> +	TP_fast_assign(
> +		__entry->dev = cur->bc_mp->m_super->s_dev;
> +		__entry->btnum = cur->bc_btnum;
> +		__entry->level = level;
> +		__entry->block_idx = block_idx;
> +		__entry->nr_blocks = nr_blocks;
> +		if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
> +			xfs_fsblock_t	fsb = be64_to_cpu(ptr->l);
> +
> +			__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsb);
> +			__entry->agbno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsb);
> +		} else {
> +			__entry->agno = cur->bc_private.a.agno;
> +			__entry->agbno = be32_to_cpu(ptr->s);
> +		}
> +		__entry->nr_records = nr_records;
> +	),
> +	TP_printk("dev %d:%d btree %s level %u block %llu/%llu fsb (%u/%u) recs %u",
> +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> +		  __entry->level,
> +		  __entry->block_idx,
> +		  __entry->nr_blocks,
> +		  __entry->agno,
> +		  __entry->agbno,
> +		  __entry->nr_records)
> +)
> +
>  #endif /* _TRACE_XFS_H */
>  
>  #undef TRACE_INCLUDE_PATH
>

Darrick J. Wong Oct. 16, 2019, 6:15 p.m. UTC | #2

On Wed, Oct 16, 2019 at 11:26:48AM -0400, Brian Foster wrote:
> On Wed, Oct 09, 2019 at 09:48:18AM -0700, Darrick J. Wong wrote:
> > From: Darrick J. Wong <[e-mail]>
> > 
> > Add a new btree function that enables us to bulk load a btree cursor.
> > This will be used by the upcoming online repair patches to generate new
> > btrees.
> > 
> > Signed-off-by: Darrick J. Wong <[e-mail]>
> > ---
> >  fs/xfs/libxfs/xfs_btree.c |  566 +++++++++++++++++++++++++++++++++++++++++++++
> >  fs/xfs/libxfs/xfs_btree.h |   43 +++
> >  fs/xfs/xfs_trace.c        |    1 
> >  fs/xfs/xfs_trace.h        |   85 +++++++
> >  4 files changed, 694 insertions(+), 1 deletion(-)
> > 
> > 
> > diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
> > index 4b06d5d86834..17b0fdb87729 100644
> > --- a/fs/xfs/libxfs/xfs_btree.c
> > +++ b/fs/xfs/libxfs/xfs_btree.c
> ...
> > @@ -5104,3 +5104,567 @@ xfs_btree_commit_ifakeroot(
> >  	cur->bc_ops = ops;
> >  	cur->bc_flags &= ~XFS_BTREE_STAGING;
> >  }
> > +
> > +/*
> > + * Bulk Loading of Staged Btrees
> > + * =============================
> > + *
> > + * This interface is used with a staged btree cursor to create a totally new
> > + * btree with a large number of records (i.e. more than what would fit in a
> > + * single block).  When the creation is complete, the new root can be linked
> > + * atomically into the filesystem by committing the staged cursor.
> > + *

[paraphrasing a conversation we had on irc]

> Thanks for the documentation. So what is the purpose behind the whole
> bulk loading thing as opposed to something like faking up an AG
> structure (i.e. AGF) somewhere and using the existing cursor mechanisms
> (or something closer to it) to copy records from one place to another?
> Is it purely a performance/efficiency tradeoff? Bulk block allocation
> issues? Transactional/atomicity issues? All (or none :P) of the above?

Prior to the v20, the online repair series created a new btree root,
committed that into wherever the root lived, and inserted records one by
one into the btree.  There were quite a few drawbacks to this method:

1. Inserting records one at a time can involve walking up the tree to
update node block pointers, which isn't terribly efficient if we're
likely going to rewrite the pointers (and relogging nodes) several more
times.

2. Inserting records one at a time tends to leave a lot of half-empty
btree blocks because when one block fills up we split it and push half
the records to the new block.  It would be nice not to explode the size
of the btrees, and it would be particularly useful if we could control
the load factor of the new btree precisely.

3. The rebuild wasn't atomic, since we were replacing the root prior to
the insert loop.  If we crashed midway through a rebuild we'd end up
with a garbage btree and no indication that it was incorrect.  That's
how the fakeroot code got started.

4. In a previous version of the repair series I tried to batch as many
insert operations into a single transaction as possible, but my
transaction reservation fullness estimation function didn't work
reliably (particularly when things got really fragmented), so I backed
off to rolling after /every/ insertion.  That works well enough, but at
a cost of a lot of transaction rolling, which means that repairs plod
along very slowly.

5. Performing an insert loop means that the btree blocks are allocated
one at a time as the btree expands.  This is suboptimal since we can
calculate the exact size of the new btree prior to building it, which
gives us the opportunity to recreate the index in a set of contiguous
blocks instead of scattering them.

6. If we crash midway through a rebuild, XFS neither cleaned up the mess
nor informed the administrator that it was necessary to re-run xfs_scrub
or xfs_repair to clean up the lost blocks.  Obviously, automatic cleanup
is a far better solution.

The first thing I decided to solve was the lack of atomicity.

For AG-rooted btrees, I thought about creating a fake xfs_buf for an AG
header buffer and extracting the root/level values after construction
completes.  That's possible, but it's risky because the fake buffer
could get logged and if the sector number matches the actual header
then it introduces buffer cache aliasing issues.

For inode-rooted btrees, one could create a fake xfs_inode with the same
i_ino as the target.  That presents the same aliasing issues as the fake
xfs_buf above.  A different strategy would be to allocate an unlinked
inode and then use the bmbt owner change (a.k.a. extent swap) to move
the mappings over.  That would work, though it has two large drawbacks:
(a) a lot of additional complexity around allocating and freeing the
temporary inode; and (b) future inode-rooted btrees such as the realtime
rmap btree would also have to implement an owner-change operation.

To fix (3), I thought it wise to have explicit fakeroot structures to
maintain a clean separation between what we're building and the rest of
the filesystem.  This also means that there's nothing on disk to clean
up if we fail at any point before we're ready to commit the new btree.

Then Dave (I think?) suggested that I  use EFIs strategically to
schedule freeing of the new btree blocks (the root commit transaction
would log EFDs to cancel them) and to schedule freeing of the old
blocks.  That solves (6), though the EFI wrangling doesn't happen for
another couple of series after this one.

He also suggested using ordered buffers to write out the new btree
blocks along with whatever logging was necessary to commit the new
btree.  It then occurred to me that xfs_repair open-codes the process of
calculating the geometry of a new btree, allocating all the blocks at
once, and writing out full btree blocks.  Somewhat annoyingly, it
features nearly the same (open-)code for all four AG btree types, which
is less maintainable than it could be.

I read through all four versions and used it to write the generic btree
bulk loading code.  For scrub I hooked that up to the "staged btree with
a fake root" stuff I'd already written, which solves (1), (2), (4), and
(5).

For xfsprogs[1], I deleted a few thousand lines of code from xfs_repair.
True, we don't reuse existing common code, but we at least get to share
new common btree code.

> This is my first pass through this so I'm mostly looking at big picture
> until I get to a point to see how these bits are used. The mechanism
> itself seems reasonable in principle, but the reason I ask is it also
> seems like there's inherent value in using more of same infrastructure
> to reconstruct a tree that we use to create one in the first place. We
> also already have primitives for things like fork swapping via the
> extent swap mechanism, etc.

"bfoster: I guess it would be nice to see that kind of make it work ->
make it fast evolution in tree"

For a while I did maintain the introduction of the bulk loading code as
separate patches against the v19 repair code, but unfortunately I
smushed them down before sending v20 to reduce the patch count, and
because I didn't want to argue with everyone over the semi-working code
that would then be replaced in the very next patch.

I could split them back out, though at a cost of having to reintroduce a
lot of hairy code in the bnobt/cntbt rebuild function to seed the free
new space btree root in order to make sure that the btree block
allocation code works properly, along with auditing the allocation paths
to make sure they don't use the old AGF or encounter other subtleties.

It'd be a lot of work considering that the v20 reconstruction code is
/much/ simpler than v19's was.  I also restructured the repair functions
to allocate one large context structure at the beginning instead of the
piecemeal way it was done onstack in v19 because stack usage was growing
close to 1k in some cases.

--D

[1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/log/?h=repair-bulk-load

> 
> Brian
> 
> > + * The first step for the caller is to construct a fake btree root structure
> > + * and a staged btree cursor.  A staging cursor contains all the geometry
> > + * information for the btree type but will fail all operations that could have
> > + * side effects in the filesystem (e.g. btree shape changes).  Regular
> > + * operations will not work unless the staging cursor is committed and becomes
> > + * a regular cursor.
> > + *
> > + * For a btree rooted in an AG header, use an xbtree_afakeroot structure.
> > + * This should be initialized to zero.  For a btree rooted in an inode fork,
> > + * use an xbtree_ifakeroot structure.  @if_fork_size field should be set to
> > + * the number of bytes available to the fork in the inode; @if_fork should
> > + * point to a freshly allocated xfs_inode_fork; and @if_format should be set
> > + * to the appropriate fork type (e.g. XFS_DINODE_FMT_BTREE).
> > + *
> > + * The next step for the caller is to initialize a struct xfs_btree_bload
> > + * context.  The @nr_records field is the number of records that are to be
> > + * loaded into the btree.  The @leaf_slack and @node_slack fields are the
> > + * number of records (or key/ptr) slots to leave empty in new btree blocks.
> > + * If a caller sets a slack value to -1, the slack value will be computed to
> > + * fill the block halfway between minrecs and maxrecs items per block.
> > + *
> > + * The number of items placed in each btree block is computed via the following
> > + * algorithm: For leaf levels, the number of items for the level is nr_records.
> > + * For node levels, the number of items for the level is the number of blocks
> > + * in the next lower level of the tree.  For each level, the desired number of
> > + * items per block is defined as:
> > + *
> > + * desired = max(minrecs, maxrecs - slack factor)
> > + *
> > + * The number of blocks for the level is defined to be:
> > + *
> > + * blocks = nr_items / desired
> > + *
> > + * Note this is rounded down so that the npb calculation below will never fall
> > + * below minrecs.  The number of items that will actually be loaded into each
> > + * btree block is defined as:
> > + *
> > + * npb =  nr_items / blocks
> > + *
> > + * Some of the leftmost blocks in the level will contain one extra record as
> > + * needed to handle uneven division.  If the number of records in any block
> > + * would exceed maxrecs for that level, blocks is incremented and npb is
> > + * recalculated.
> > + *
> > + * In other words, we compute the number of blocks needed to satisfy a given
> > + * loading level, then spread the items as evenly as possible.
> > + *
> > + * To complete this step, call xfs_btree_bload_compute_geometry, which uses
> > + * those settings to compute the height of the btree and the number of blocks
> > + * that will be needed to construct the btree.  These values are stored in the
> > + * @btree_height and @nr_blocks fields.
> > + *
> > + * At this point, the caller must allocate @nr_blocks blocks and save them for
> > + * later.  If space is to be allocated transactionally, the staging cursor
> > + * must be deleted before and recreated after, which is why computing the
> > + * geometry is a separate step.
> > + *
> > + * The fourth step in the bulk loading process is to set the function pointers
> > + * in the bload context structure.  @get_data will be called for each record
> > + * that will be loaded into the btree; it should set the cursor's bc_rec
> > + * field, which will be converted to on-disk format and copied into the
> > + * appropriate record slot.  @alloc_block should supply one of the blocks
> > + * allocated in the previous step.  For btrees which are rooted in an inode
> > + * fork, @iroot_size is called to compute the size of the incore btree root
> > + * block.  Call xfs_btree_bload to start constructing the btree.
> > + *
> > + * The final step is to commit the staging cursor, which logs the new btree
> > + * root and turns the btree into a regular btree cursor, and free the fake
> > + * roots.
> > + */
> > +
> > +/*
> > + * Put a btree block that we're loading onto the ordered list and release it.
> > + * The btree blocks will be written when the final transaction swapping the
> > + * btree roots is committed.
> > + */
> > +static void
> > +xfs_btree_bload_drop_buf(
> > +	struct xfs_trans	*tp,
> > +	struct xfs_buf		**bpp)
> > +{
> > +	if (*bpp == NULL)
> > +		return;
> > +
> > +	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_BTREE_BUF);
> > +	xfs_trans_ordered_buf(tp, *bpp);
> > +	xfs_trans_brelse(tp, *bpp);
> > +	*bpp = NULL;
> > +}
> > +
> > +/* Allocate and initialize one btree block for bulk loading. */
> > +STATIC int
> > +xfs_btree_bload_prep_block(
> > +	struct xfs_btree_cur		*cur,
> > +	struct xfs_btree_bload		*bbl,
> > +	unsigned int			level,
> > +	unsigned int			nr_this_block,
> > +	union xfs_btree_ptr		*ptrp,
> > +	struct xfs_buf			**bpp,
> > +	struct xfs_btree_block		**blockp,
> > +	void				*priv)
> > +{
> > +	union xfs_btree_ptr		new_ptr;
> > +	struct xfs_buf			*new_bp;
> > +	struct xfs_btree_block		*new_block;
> > +	int				ret;
> > +
> > +	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
> > +	    level == cur->bc_nlevels - 1) {
> > +		struct xfs_ifork	*ifp = cur->bc_private.b.ifake->if_fork;
> > +		size_t			new_size;
> > +
> > +		/* Allocate a new incore btree root block. */
> > +		new_size = bbl->iroot_size(cur, nr_this_block, priv);
> > +		ifp->if_broot = kmem_zalloc(new_size, 0);
> > +		ifp->if_broot_bytes = (int)new_size;
> > +		ifp->if_flags |= XFS_IFBROOT;
> > +
> > +		/* Initialize it and send it out. */
> > +		xfs_btree_init_block_int(cur->bc_mp, ifp->if_broot,
> > +				XFS_BUF_DADDR_NULL, cur->bc_btnum, level,
> > +				nr_this_block, cur->bc_private.b.ip->i_ino,
> > +				cur->bc_flags);
> > +
> > +		*bpp = NULL;
> > +		*blockp = ifp->if_broot;
> > +		xfs_btree_set_ptr_null(cur, ptrp);
> > +		return 0;
> > +	}
> > +
> > +	/* Allocate a new leaf block. */
> > +	ret = bbl->alloc_block(cur, &new_ptr, priv);
> > +	if (ret)
> > +		return ret;
> > +
> > +	ASSERT(!xfs_btree_ptr_is_null(cur, &new_ptr));
> > +
> > +	ret = xfs_btree_get_buf_block(cur, &new_ptr, &new_block, &new_bp);
> > +	if (ret)
> > +		return ret;
> > +
> > +	/* Initialize the btree block. */
> > +	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
> > +	if (*blockp)
> > +		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
> > +	xfs_btree_set_sibling(cur, new_block, ptrp, XFS_BB_LEFTSIB);
> > +	xfs_btree_set_numrecs(new_block, nr_this_block);
> > +
> > +	/* Release the old block and set the out parameters. */
> > +	xfs_btree_bload_drop_buf(cur->bc_tp, bpp);
> > +	*blockp = new_block;
> > +	*bpp = new_bp;
> > +	xfs_btree_copy_ptrs(cur, ptrp, &new_ptr, 1);
> > +	return 0;
> > +}
> > +
> > +/* Load one leaf block. */
> > +STATIC int
> > +xfs_btree_bload_leaf(
> > +	struct xfs_btree_cur		*cur,
> > +	unsigned int			recs_this_block,
> > +	xfs_btree_bload_get_fn		get_data,
> > +	struct xfs_btree_block		*block,
> > +	void				*priv)
> > +{
> > +	unsigned int			j;
> > +	int				ret;
> > +
> > +	/* Fill the leaf block with records. */
> > +	for (j = 1; j <= recs_this_block; j++) {
> > +		union xfs_btree_rec	*block_recs;
> > +
> > +		ret = get_data(cur, priv);
> > +		if (ret)
> > +			return ret;
> > +		block_recs = xfs_btree_rec_addr(cur, j, block);
> > +		cur->bc_ops->init_rec_from_cur(cur, block_recs);
> > +	}
> > +
> > +	return 0;
> > +}
> > +
> > +/* Load one node block. */
> > +STATIC int
> > +xfs_btree_bload_node(
> > +	struct xfs_btree_cur	*cur,
> > +	unsigned int		recs_this_block,
> > +	union xfs_btree_ptr	*child_ptr,
> > +	struct xfs_btree_block	*block)
> > +{
> > +	unsigned int		j;
> > +	int			ret;
> > +
> > +	/* Fill the node block with keys and pointers. */
> > +	for (j = 1; j <= recs_this_block; j++) {
> > +		union xfs_btree_key	child_key;
> > +		union xfs_btree_ptr	*block_ptr;
> > +		union xfs_btree_key	*block_key;
> > +		struct xfs_btree_block	*child_block;
> > +		struct xfs_buf		*child_bp;
> > +
> > +		ASSERT(!xfs_btree_ptr_is_null(cur, child_ptr));
> > +
> > +		ret = xfs_btree_get_buf_block(cur, child_ptr, &child_block,
> > +				&child_bp);
> > +		if (ret)
> > +			return ret;
> > +
> > +		xfs_btree_get_keys(cur, child_block, &child_key);
> > +
> > +		block_ptr = xfs_btree_ptr_addr(cur, j, block);
> > +		xfs_btree_copy_ptrs(cur, block_ptr, child_ptr, 1);
> > +
> > +		block_key = xfs_btree_key_addr(cur, j, block);
> > +		xfs_btree_copy_keys(cur, block_key, &child_key, 1);
> > +
> > +		xfs_btree_get_sibling(cur, child_block, child_ptr,
> > +				XFS_BB_RIGHTSIB);
> > +		xfs_trans_brelse(cur->bc_tp, child_bp);
> > +	}
> > +
> > +	return 0;
> > +}
> > +
> > +/*
> > + * Compute the maximum number of records (or keyptrs) per block that we want to
> > + * install at this level in the btree.  Caller is responsible for having set
> > + * @cur->bc_private.b.forksize to the desired fork size, if appropriate.
> > + */
> > +STATIC unsigned int
> > +xfs_btree_bload_max_npb(
> > +	struct xfs_btree_cur	*cur,
> > +	struct xfs_btree_bload	*bbl,
> > +	unsigned int		level)
> > +{
> > +	unsigned int		ret;
> > +
> > +	if (level == cur->bc_nlevels - 1 && cur->bc_ops->get_dmaxrecs)
> > +		return cur->bc_ops->get_dmaxrecs(cur, level);
> > +
> > +	ret = cur->bc_ops->get_maxrecs(cur, level);
> > +	if (level == 0)
> > +		ret -= bbl->leaf_slack;
> > +	else
> > +		ret -= bbl->node_slack;
> > +	return ret;
> > +}
> > +
> > +/*
> > + * Compute the desired number of records (or keyptrs) per block that we want to
> > + * install at this level in the btree, which must be somewhere between minrecs
> > + * and max_npb.  The caller is free to install fewer records per block.
> > + */
> > +STATIC unsigned int
> > +xfs_btree_bload_desired_npb(
> > +	struct xfs_btree_cur	*cur,
> > +	struct xfs_btree_bload	*bbl,
> > +	unsigned int		level)
> > +{
> > +	unsigned int		npb = xfs_btree_bload_max_npb(cur, bbl, level);
> > +
> > +	/* Root blocks are not subject to minrecs rules. */
> > +	if (level == cur->bc_nlevels - 1)
> > +		return max(1U, npb);
> > +
> > +	return max_t(unsigned int, cur->bc_ops->get_minrecs(cur, level), npb);
> > +}
> > +
> > +/*
> > + * Compute the number of records to be stored in each block at this level and
> > + * the number of blocks for this level.  For leaf levels, we must populate an
> > + * empty root block even if there are no records, so we have to have at least
> > + * one block.
> > + */
> > +STATIC void
> > +xfs_btree_bload_level_geometry(
> > +	struct xfs_btree_cur	*cur,
> > +	struct xfs_btree_bload	*bbl,
> > +	unsigned int		level,
> > +	uint64_t		nr_this_level,
> > +	unsigned int		*avg_per_block,
> > +	uint64_t		*blocks,
> > +	uint64_t		*blocks_with_extra)
> > +{
> > +	uint64_t		npb;
> > +	uint64_t		dontcare;
> > +	unsigned int		desired_npb;
> > +	unsigned int		maxnr;
> > +
> > +	maxnr = cur->bc_ops->get_maxrecs(cur, level);
> > +
> > +	/*
> > +	 * Compute the number of blocks we need to fill each block with the
> > +	 * desired number of records/keyptrs per block.  Because desired_npb
> > +	 * could be minrecs, we use regular integer division (which rounds
> > +	 * the block count down) so that in the next step the effective # of
> > +	 * items per block will never be less than desired_npb.
> > +	 */
> > +	desired_npb = xfs_btree_bload_desired_npb(cur, bbl, level);
> > +	*blocks = div64_u64_rem(nr_this_level, desired_npb, &dontcare);
> > +	*blocks = max(1ULL, *blocks);
> > +
> > +	/*
> > +	 * Compute the number of records that we will actually put in each
> > +	 * block, assuming that we want to spread the records evenly between
> > +	 * the blocks.  Take care that the effective # of items per block (npb)
> > +	 * won't exceed maxrecs even for the blocks that get an extra record,
> > +	 * since desired_npb could be maxrecs, and in the previous step we
> > +	 * rounded the block count down.
> > +	 */
> > +	npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > +	if (npb > maxnr || (npb == maxnr && *blocks_with_extra > 0)) {
> > +		(*blocks)++;
> > +		npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > +	}
> > +
> > +	*avg_per_block = min_t(uint64_t, npb, nr_this_level);
> > +
> > +	trace_xfs_btree_bload_level_geometry(cur, level, nr_this_level,
> > +			*avg_per_block, desired_npb, *blocks,
> > +			*blocks_with_extra);
> > +}
> > +
> > +/*
> > + * Ensure a slack value is appropriate for the btree.
> > + *
> > + * If the slack value is negative, set slack so that we fill the block to
> > + * halfway between minrecs and maxrecs.  Make sure the slack is never so large
> > + * that we can underflow minrecs.
> > + */
> > +static void
> > +xfs_btree_bload_ensure_slack(
> > +	struct xfs_btree_cur	*cur,
> > +	int			*slack,
> > +	int			level)
> > +{
> > +	int			maxr;
> > +	int			minr;
> > +
> > +	/*
> > +	 * We only care about slack for btree blocks, so set the btree nlevels
> > +	 * to 3 so that level 0 is a leaf block and level 1 is a node block.
> > +	 * Avoid straying into inode roots, since we don't do slack there.
> > +	 */
> > +	cur->bc_nlevels = 3;
> > +	maxr = cur->bc_ops->get_maxrecs(cur, level);
> > +	minr = cur->bc_ops->get_minrecs(cur, level);
> > +
> > +	/*
> > +	 * If slack is negative, automatically set slack so that we load the
> > +	 * btree block approximately halfway between minrecs and maxrecs.
> > +	 * Generally, this will net us 75% loading.
> > +	 */
> > +	if (*slack < 0)
> > +		*slack = maxr - ((maxr + minr) >> 1);
> > +
> > +	*slack = min(*slack, maxr - minr);
> > +}
> > +
> > +/*
> > + * Prepare a btree cursor for a bulk load operation by computing the geometry
> > + * fields in @bbl.  Caller must ensure that the btree cursor is a staging
> > + * cursor.  This function can be called multiple times.
> > + */
> > +int
> > +xfs_btree_bload_compute_geometry(
> > +	struct xfs_btree_cur	*cur,
> > +	struct xfs_btree_bload	*bbl,
> > +	uint64_t		nr_records)
> > +{
> > +	uint64_t		nr_blocks = 0;
> > +	uint64_t		nr_this_level;
> > +
> > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > +
> > +	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
> > +	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
> > +
> > +	bbl->nr_records = nr_this_level = nr_records;
> > +	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
> > +		uint64_t	level_blocks;
> > +		uint64_t	dontcare64;
> > +		unsigned int	level = cur->bc_nlevels - 1;
> > +		unsigned int	avg_per_block;
> > +
> > +		/*
> > +		 * If all the things we want to store at this level would fit
> > +		 * in a single root block, then we have our btree root and are
> > +		 * done.  Note that bmap btrees do not allow records in the
> > +		 * root.
> > +		 */
> > +		if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) || level != 0) {
> > +			xfs_btree_bload_level_geometry(cur, bbl, level,
> > +					nr_this_level, &avg_per_block,
> > +					&level_blocks, &dontcare64);
> > +			if (nr_this_level <= avg_per_block) {
> > +				nr_blocks++;
> > +				break;
> > +			}
> > +		}
> > +
> > +		/*
> > +		 * Otherwise, we have to store all the records for this level
> > +		 * in blocks and therefore need another level of btree to point
> > +		 * to those blocks.  Increase the number of levels and
> > +		 * recompute the number of records we can store at this level
> > +		 * because that can change depending on whether or not a level
> > +		 * is the root level.
> > +		 */
> > +		cur->bc_nlevels++;
> > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > +				&avg_per_block, &level_blocks, &dontcare64);
> > +		nr_blocks += level_blocks;
> > +		nr_this_level = level_blocks;
> > +	}
> > +
> > +	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
> > +		return -EOVERFLOW;
> > +
> > +	bbl->btree_height = cur->bc_nlevels;
> > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
> > +		bbl->nr_blocks = nr_blocks - 1;
> > +	else
> > +		bbl->nr_blocks = nr_blocks;
> > +	return 0;
> > +}
> > +
> > +/*
> > + * Bulk load a btree.
> > + *
> > + * Load @bbl->nr_records quantity of records into a btree using the supplied
> > + * empty and staging btree cursor @cur and a @bbl that has been filled out by
> > + * the xfs_btree_bload_compute_geometry function.
> > + *
> > + * The @bbl->get_data function must populate the cursor's bc_rec every time it
> > + * is called.  The @bbl->alloc_block function will be used to allocate new
> > + * btree blocks.  @priv is passed to both functions.
> > + *
> > + * Caller must ensure that @cur is a staging cursor.  Any existing btree rooted
> > + * in the fakeroot will be lost, so do not call this function twice.
> > + */
> > +int
> > +xfs_btree_bload(
> > +	struct xfs_btree_cur		*cur,
> > +	struct xfs_btree_bload		*bbl,
> > +	void				*priv)
> > +{
> > +	union xfs_btree_ptr		child_ptr;
> > +	union xfs_btree_ptr		ptr;
> > +	struct xfs_buf			*bp = NULL;
> > +	struct xfs_btree_block		*block = NULL;
> > +	uint64_t			nr_this_level = bbl->nr_records;
> > +	uint64_t			blocks;
> > +	uint64_t			i;
> > +	uint64_t			blocks_with_extra;
> > +	uint64_t			total_blocks = 0;
> > +	unsigned int			avg_per_block;
> > +	unsigned int			level = 0;
> > +	int				ret;
> > +
> > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > +
> > +	cur->bc_nlevels = bbl->btree_height;
> > +	xfs_btree_set_ptr_null(cur, &child_ptr);
> > +	xfs_btree_set_ptr_null(cur, &ptr);
> > +
> > +	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > +			&avg_per_block, &blocks, &blocks_with_extra);
> > +
> > +	/* Load each leaf block. */
> > +	for (i = 0; i < blocks; i++) {
> > +		unsigned int		nr_this_block = avg_per_block;
> > +
> > +		if (i < blocks_with_extra)
> > +			nr_this_block++;
> > +
> > +		ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > +				nr_this_block, &ptr, &bp, &block, priv);
> > +		if (ret)
> > +			return ret;
> > +
> > +		trace_xfs_btree_bload_block(cur, level, i, blocks, &ptr,
> > +				nr_this_block);
> > +
> > +		ret = xfs_btree_bload_leaf(cur, nr_this_block, bbl->get_data,
> > +				block, priv);
> > +		if (ret)
> > +			goto out;
> > +
> > +		/* Record the leftmost pointer to start the next level. */
> > +		if (i == 0)
> > +			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
> > +	}
> > +	total_blocks += blocks;
> > +	xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > +
> > +	/* Populate the internal btree nodes. */
> > +	for (level = 1; level < cur->bc_nlevels; level++) {
> > +		union xfs_btree_ptr	first_ptr;
> > +
> > +		nr_this_level = blocks;
> > +		block = NULL;
> > +		xfs_btree_set_ptr_null(cur, &ptr);
> > +
> > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > +				&avg_per_block, &blocks, &blocks_with_extra);
> > +
> > +		/* Load each node block. */
> > +		for (i = 0; i < blocks; i++) {
> > +			unsigned int	nr_this_block = avg_per_block;
> > +
> > +			if (i < blocks_with_extra)
> > +				nr_this_block++;
> > +
> > +			ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > +					nr_this_block, &ptr, &bp, &block,
> > +					priv);
> > +			if (ret)
> > +				return ret;
> > +
> > +			trace_xfs_btree_bload_block(cur, level, i, blocks,
> > +					&ptr, nr_this_block);
> > +
> > +			ret = xfs_btree_bload_node(cur, nr_this_block,
> > +					&child_ptr, block);
> > +			if (ret)
> > +				goto out;
> > +
> > +			/*
> > +			 * Record the leftmost pointer to start the next level.
> > +			 */
> > +			if (i == 0)
> > +				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
> > +		}
> > +		total_blocks += blocks;
> > +		xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > +		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
> > +	}
> > +
> > +	/* Initialize the new root. */
> > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
> > +		ASSERT(xfs_btree_ptr_is_null(cur, &ptr));
> > +		cur->bc_private.b.ifake->if_levels = cur->bc_nlevels;
> > +		cur->bc_private.b.ifake->if_blocks = total_blocks - 1;
> > +	} else {
> > +		cur->bc_private.a.afake->af_root = be32_to_cpu(ptr.s);
> > +		cur->bc_private.a.afake->af_levels = cur->bc_nlevels;
> > +		cur->bc_private.a.afake->af_blocks = total_blocks;
> > +	}
> > +out:
> > +	if (bp)
> > +		xfs_trans_brelse(cur->bc_tp, bp);
> > +	return ret;
> > +}
> > diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
> > index a17becb72ab8..5c6992a04ea2 100644
> > --- a/fs/xfs/libxfs/xfs_btree.h
> > +++ b/fs/xfs/libxfs/xfs_btree.h
> > @@ -582,4 +582,47 @@ void xfs_btree_stage_ifakeroot(struct xfs_btree_cur *cur,
> >  void xfs_btree_commit_ifakeroot(struct xfs_btree_cur *cur, int whichfork,
> >  		const struct xfs_btree_ops *ops);
> >  
> > +typedef int (*xfs_btree_bload_get_fn)(struct xfs_btree_cur *cur, void *priv);
> > +typedef int (*xfs_btree_bload_alloc_block_fn)(struct xfs_btree_cur *cur,
> > +		union xfs_btree_ptr *ptr, void *priv);
> > +typedef size_t (*xfs_btree_bload_iroot_size_fn)(struct xfs_btree_cur *cur,
> > +		unsigned int nr_this_level, void *priv);
> > +
> > +/* Bulk loading of staged btrees. */
> > +struct xfs_btree_bload {
> > +	/* Function to store a record in the cursor. */
> > +	xfs_btree_bload_get_fn		get_data;
> > +
> > +	/* Function to allocate a block for the btree. */
> > +	xfs_btree_bload_alloc_block_fn	alloc_block;
> > +
> > +	/* Function to compute the size of the in-core btree root block. */
> > +	xfs_btree_bload_iroot_size_fn	iroot_size;
> > +
> > +	/* Number of records the caller wants to store. */
> > +	uint64_t			nr_records;
> > +
> > +	/* Number of btree blocks needed to store those records. */
> > +	uint64_t			nr_blocks;
> > +
> > +	/*
> > +	 * Number of free records to leave in each leaf block.  If this (or
> > +	 * any of the slack values) are negative, this will be computed to
> > +	 * be halfway between maxrecs and minrecs.  This typically leaves the
> > +	 * block 75% full.
> > +	 */
> > +	int				leaf_slack;
> > +
> > +	/* Number of free keyptrs to leave in each node block. */
> > +	int				node_slack;
> > +
> > +	/* Computed btree height. */
> > +	unsigned int			btree_height;
> > +};
> > +
> > +int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
> > +		struct xfs_btree_bload *bbl, uint64_t nr_records);
> > +int xfs_btree_bload(struct xfs_btree_cur *cur, struct xfs_btree_bload *bbl,
> > +		void *priv);
> > +
> >  #endif	/* __XFS_BTREE_H__ */
> > diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
> > index bc85b89f88ca..9b5e58a92381 100644
> > --- a/fs/xfs/xfs_trace.c
> > +++ b/fs/xfs/xfs_trace.c
> > @@ -6,6 +6,7 @@
> >  #include "xfs.h"
> >  #include "xfs_fs.h"
> >  #include "xfs_shared.h"
> > +#include "xfs_bit.h"
> >  #include "xfs_format.h"
> >  #include "xfs_log_format.h"
> >  #include "xfs_trans_resv.h"
> > diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
> > index a78055521fcd..6d7ba64b7a0f 100644
> > --- a/fs/xfs/xfs_trace.h
> > +++ b/fs/xfs/xfs_trace.h
> > @@ -35,6 +35,7 @@ struct xfs_icreate_log;
> >  struct xfs_owner_info;
> >  struct xfs_trans_res;
> >  struct xfs_inobt_rec_incore;
> > +union xfs_btree_ptr;
> >  
> >  DECLARE_EVENT_CLASS(xfs_attr_list_class,
> >  	TP_PROTO(struct xfs_attr_list_context *ctx),
> > @@ -3670,6 +3671,90 @@ TRACE_EVENT(xfs_btree_commit_ifakeroot,
> >  		  __entry->blocks)
> >  )
> >  
> > +TRACE_EVENT(xfs_btree_bload_level_geometry,
> > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > +		 uint64_t nr_this_level, unsigned int nr_per_block,
> > +		 unsigned int desired_npb, uint64_t blocks,
> > +		 uint64_t blocks_with_extra),
> > +	TP_ARGS(cur, level, nr_this_level, nr_per_block, desired_npb, blocks,
> > +		blocks_with_extra),
> > +	TP_STRUCT__entry(
> > +		__field(dev_t, dev)
> > +		__field(xfs_btnum_t, btnum)
> > +		__field(unsigned int, level)
> > +		__field(unsigned int, nlevels)
> > +		__field(uint64_t, nr_this_level)
> > +		__field(unsigned int, nr_per_block)
> > +		__field(unsigned int, desired_npb)
> > +		__field(unsigned long long, blocks)
> > +		__field(unsigned long long, blocks_with_extra)
> > +	),
> > +	TP_fast_assign(
> > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > +		__entry->btnum = cur->bc_btnum;
> > +		__entry->level = level;
> > +		__entry->nlevels = cur->bc_nlevels;
> > +		__entry->nr_this_level = nr_this_level;
> > +		__entry->nr_per_block = nr_per_block;
> > +		__entry->desired_npb = desired_npb;
> > +		__entry->blocks = blocks;
> > +		__entry->blocks_with_extra = blocks_with_extra;
> > +	),
> > +	TP_printk("dev %d:%d btree %s level %u/%u nr_this_level %llu nr_per_block %u desired_npb %u blocks %llu blocks_with_extra %llu",
> > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > +		  __entry->level,
> > +		  __entry->nlevels,
> > +		  __entry->nr_this_level,
> > +		  __entry->nr_per_block,
> > +		  __entry->desired_npb,
> > +		  __entry->blocks,
> > +		  __entry->blocks_with_extra)
> > +)
> > +
> > +TRACE_EVENT(xfs_btree_bload_block,
> > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > +		 uint64_t block_idx, uint64_t nr_blocks,
> > +		 union xfs_btree_ptr *ptr, unsigned int nr_records),
> > +	TP_ARGS(cur, level, block_idx, nr_blocks, ptr, nr_records),
> > +	TP_STRUCT__entry(
> > +		__field(dev_t, dev)
> > +		__field(xfs_btnum_t, btnum)
> > +		__field(unsigned int, level)
> > +		__field(unsigned long long, block_idx)
> > +		__field(unsigned long long, nr_blocks)
> > +		__field(xfs_agnumber_t, agno)
> > +		__field(xfs_agblock_t, agbno)
> > +		__field(unsigned int, nr_records)
> > +	),
> > +	TP_fast_assign(
> > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > +		__entry->btnum = cur->bc_btnum;
> > +		__entry->level = level;
> > +		__entry->block_idx = block_idx;
> > +		__entry->nr_blocks = nr_blocks;
> > +		if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
> > +			xfs_fsblock_t	fsb = be64_to_cpu(ptr->l);
> > +
> > +			__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsb);
> > +			__entry->agbno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsb);
> > +		} else {
> > +			__entry->agno = cur->bc_private.a.agno;
> > +			__entry->agbno = be32_to_cpu(ptr->s);
> > +		}
> > +		__entry->nr_records = nr_records;
> > +	),
> > +	TP_printk("dev %d:%d btree %s level %u block %llu/%llu fsb (%u/%u) recs %u",
> > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > +		  __entry->level,
> > +		  __entry->block_idx,
> > +		  __entry->nr_blocks,
> > +		  __entry->agno,
> > +		  __entry->agbno,
> > +		  __entry->nr_records)
> > +)
> > +
> >  #endif /* _TRACE_XFS_H */
> >  
> >  #undef TRACE_INCLUDE_PATH
> >

Brian Foster Oct. 16, 2019, 9:07 p.m. UTC | #3

On Wed, Oct 16, 2019 at 11:15:02AM -0700, Darrick J. Wong wrote:
> On Wed, Oct 16, 2019 at 11:26:48AM -0400, Brian Foster wrote:
> > On Wed, Oct 09, 2019 at 09:48:18AM -0700, Darrick J. Wong wrote:
> > > From: Darrick J. Wong <[e-mail]>
> > > 
> > > Add a new btree function that enables us to bulk load a btree cursor.
> > > This will be used by the upcoming online repair patches to generate new
> > > btrees.
> > > 
> > > Signed-off-by: Darrick J. Wong <[e-mail]>
> > > ---
> > >  fs/xfs/libxfs/xfs_btree.c |  566 +++++++++++++++++++++++++++++++++++++++++++++
> > >  fs/xfs/libxfs/xfs_btree.h |   43 +++
> > >  fs/xfs/xfs_trace.c        |    1 
> > >  fs/xfs/xfs_trace.h        |   85 +++++++
> > >  4 files changed, 694 insertions(+), 1 deletion(-)
> > > 
> > > 
> > > diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
> > > index 4b06d5d86834..17b0fdb87729 100644
> > > --- a/fs/xfs/libxfs/xfs_btree.c
> > > +++ b/fs/xfs/libxfs/xfs_btree.c
> > ...
> > > @@ -5104,3 +5104,567 @@ xfs_btree_commit_ifakeroot(
> > >  	cur->bc_ops = ops;
> > >  	cur->bc_flags &= ~XFS_BTREE_STAGING;
> > >  }
> > > +
> > > +/*
> > > + * Bulk Loading of Staged Btrees
> > > + * =============================
> > > + *
> > > + * This interface is used with a staged btree cursor to create a totally new
> > > + * btree with a large number of records (i.e. more than what would fit in a
> > > + * single block).  When the creation is complete, the new root can be linked
> > > + * atomically into the filesystem by committing the staged cursor.
> > > + *
> 
> [paraphrasing a conversation we had on irc]
> 
> > Thanks for the documentation. So what is the purpose behind the whole
> > bulk loading thing as opposed to something like faking up an AG
> > structure (i.e. AGF) somewhere and using the existing cursor mechanisms
> > (or something closer to it) to copy records from one place to another?
> > Is it purely a performance/efficiency tradeoff? Bulk block allocation
> > issues? Transactional/atomicity issues? All (or none :P) of the above?
> 
> Prior to the v20, the online repair series created a new btree root,
> committed that into wherever the root lived, and inserted records one by
> one into the btree.  There were quite a few drawbacks to this method:
> 
> 1. Inserting records one at a time can involve walking up the tree to
> update node block pointers, which isn't terribly efficient if we're
> likely going to rewrite the pointers (and relogging nodes) several more
> times.
> 
> 2. Inserting records one at a time tends to leave a lot of half-empty
> btree blocks because when one block fills up we split it and push half
> the records to the new block.  It would be nice not to explode the size
> of the btrees, and it would be particularly useful if we could control
> the load factor of the new btree precisely.
> 

Interesting... this is a trait the traditional btree update paths share
though, right?

> 3. The rebuild wasn't atomic, since we were replacing the root prior to
> the insert loop.  If we crashed midway through a rebuild we'd end up
> with a garbage btree and no indication that it was incorrect.  That's
> how the fakeroot code got started.
> 

Indeed, though this seems more related to the anchoring (i.e. fake root)
approach than bulk vs. iterative construction.

> 4. In a previous version of the repair series I tried to batch as many
> insert operations into a single transaction as possible, but my
> transaction reservation fullness estimation function didn't work
> reliably (particularly when things got really fragmented), so I backed
> off to rolling after /every/ insertion.  That works well enough, but at
> a cost of a lot of transaction rolling, which means that repairs plod
> along very slowly.
> 
> 5. Performing an insert loop means that the btree blocks are allocated
> one at a time as the btree expands.  This is suboptimal since we can
> calculate the exact size of the new btree prior to building it, which
> gives us the opportunity to recreate the index in a set of contiguous
> blocks instead of scattering them.
> 

Yep, FWIW it sounds like most of these tradeoffs are around
performance/efficiency. 

> 6. If we crash midway through a rebuild, XFS neither cleaned up the mess
> nor informed the administrator that it was necessary to re-run xfs_scrub
> or xfs_repair to clean up the lost blocks.  Obviously, automatic cleanup
> is a far better solution.
> 

Similar to above, I think this kind of depends more on how/where to
anchor an in-progress tree as opposed to what level records are copied
at.

> The first thing I decided to solve was the lack of atomicity.
> 
> For AG-rooted btrees, I thought about creating a fake xfs_buf for an AG
> header buffer and extracting the root/level values after construction
> completes.  That's possible, but it's risky because the fake buffer
> could get logged and if the sector number matches the actual header
> then it introduces buffer cache aliasing issues.
> 
> For inode-rooted btrees, one could create a fake xfs_inode with the same
> i_ino as the target.  That presents the same aliasing issues as the fake
> xfs_buf above.  A different strategy would be to allocate an unlinked
> inode and then use the bmbt owner change (a.k.a. extent swap) to move
> the mappings over.  That would work, though it has two large drawbacks:
> (a) a lot of additional complexity around allocating and freeing the
> temporary inode; and (b) future inode-rooted btrees such as the realtime
> rmap btree would also have to implement an owner-change operation.
> 

I was wondering more along the lines of having an actual anchor
somewhere. E.g., think of it as a temporary/inaccessible location of a
legitimate on-disk structure as opposed to a fake object in memory
somewhere. A hidden/internal repair inode or some such, perhaps. I'm
sure there's new code/complexity that would come around with that, but I
think that's going to be unavoidable to some degree for an online repair
mechanism. ;)

Note that this is all just handwaving on my part and still without full
context as to how things are currently anchored, made atomic, etc. I'm
primarily trying to understand the design reasoning based on the high
level description.

> To fix (3), I thought it wise to have explicit fakeroot structures to
> maintain a clean separation between what we're building and the rest of
> the filesystem.  This also means that there's nothing on disk to clean
> up if we fail at any point before we're ready to commit the new btree.
> 

Hmm.. so this approach facilites a tree reconstruction in a single open
transaction? If so, I suppose I could see some functional advantages to
that.

> Then Dave (I think?) suggested that I  use EFIs strategically to
> schedule freeing of the new btree blocks (the root commit transaction
> would log EFDs to cancel them) and to schedule freeing of the old
> blocks.  That solves (6), though the EFI wrangling doesn't happen for
> another couple of series after this one.
> 

Hm, Ok... so new btree block allocation(s?) in the same transaction as
an EFI, to be processed on recovery if we crash, otherwise cancelled
with an EFD on construction completion..?

> He also suggested using ordered buffers to write out the new btree
> blocks along with whatever logging was necessary to commit the new
> btree.  It then occurred to me that xfs_repair open-codes the process of
> calculating the geometry of a new btree, allocating all the blocks at
> once, and writing out full btree blocks.  Somewhat annoyingly, it
> features nearly the same (open-)code for all four AG btree types, which
> is less maintainable than it could be.
> 
> I read through all four versions and used it to write the generic btree
> bulk loading code.  For scrub I hooked that up to the "staged btree with
> a fake root" stuff I'd already written, which solves (1), (2), (4), and
> (5).
> 
> For xfsprogs[1], I deleted a few thousand lines of code from xfs_repair.
> True, we don't reuse existing common code, but we at least get to share
> new common btree code.
> 

Yeah, the xfsprogs work certainly makes sense. Part of the reason I ask
about this is the tradeoff of having multiple avenues to construct a
tree in the kernel codebase.

> > This is my first pass through this so I'm mostly looking at big picture
> > until I get to a point to see how these bits are used. The mechanism
> > itself seems reasonable in principle, but the reason I ask is it also
> > seems like there's inherent value in using more of same infrastructure
> > to reconstruct a tree that we use to create one in the first place. We
> > also already have primitives for things like fork swapping via the
> > extent swap mechanism, etc.
> 
> "bfoster: I guess it would be nice to see that kind of make it work ->
> make it fast evolution in tree"
> 
> For a while I did maintain the introduction of the bulk loading code as
> separate patches against the v19 repair code, but unfortunately I
> smushed them down before sending v20 to reduce the patch count, and
> because I didn't want to argue with everyone over the semi-working code
> that would then be replaced in the very next patch.
> 

That's not quite what I meant... The approach you've taken makes sense
to me for an implementation presented in a single series. I was more
thinking that at the point where it was determined the implementation
was going to change so drastically, after so many iterations it might
have been useful to see the v19 approach merged in an experimental form
and then reworked upstream from there. Now that the new approach is
implemented, I agree it's probably not worth reinserting the old
approach at this point just to switch it out.

Thanks for the breakdown...

Brian

> I could split them back out, though at a cost of having to reintroduce a
> lot of hairy code in the bnobt/cntbt rebuild function to seed the free
> new space btree root in order to make sure that the btree block
> allocation code works properly, along with auditing the allocation paths
> to make sure they don't use the old AGF or encounter other subtleties.
> 
> It'd be a lot of work considering that the v20 reconstruction code is
> /much/ simpler than v19's was.  I also restructured the repair functions
> to allocate one large context structure at the beginning instead of the
> piecemeal way it was done onstack in v19 because stack usage was growing
> close to 1k in some cases.
> 
> --D
> 
> [1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/log/?h=repair-bulk-load
> 
> > 
> > Brian
> > 
> > > + * The first step for the caller is to construct a fake btree root structure
> > > + * and a staged btree cursor.  A staging cursor contains all the geometry
> > > + * information for the btree type but will fail all operations that could have
> > > + * side effects in the filesystem (e.g. btree shape changes).  Regular
> > > + * operations will not work unless the staging cursor is committed and becomes
> > > + * a regular cursor.
> > > + *
> > > + * For a btree rooted in an AG header, use an xbtree_afakeroot structure.
> > > + * This should be initialized to zero.  For a btree rooted in an inode fork,
> > > + * use an xbtree_ifakeroot structure.  @if_fork_size field should be set to
> > > + * the number of bytes available to the fork in the inode; @if_fork should
> > > + * point to a freshly allocated xfs_inode_fork; and @if_format should be set
> > > + * to the appropriate fork type (e.g. XFS_DINODE_FMT_BTREE).
> > > + *
> > > + * The next step for the caller is to initialize a struct xfs_btree_bload
> > > + * context.  The @nr_records field is the number of records that are to be
> > > + * loaded into the btree.  The @leaf_slack and @node_slack fields are the
> > > + * number of records (or key/ptr) slots to leave empty in new btree blocks.
> > > + * If a caller sets a slack value to -1, the slack value will be computed to
> > > + * fill the block halfway between minrecs and maxrecs items per block.
> > > + *
> > > + * The number of items placed in each btree block is computed via the following
> > > + * algorithm: For leaf levels, the number of items for the level is nr_records.
> > > + * For node levels, the number of items for the level is the number of blocks
> > > + * in the next lower level of the tree.  For each level, the desired number of
> > > + * items per block is defined as:
> > > + *
> > > + * desired = max(minrecs, maxrecs - slack factor)
> > > + *
> > > + * The number of blocks for the level is defined to be:
> > > + *
> > > + * blocks = nr_items / desired
> > > + *
> > > + * Note this is rounded down so that the npb calculation below will never fall
> > > + * below minrecs.  The number of items that will actually be loaded into each
> > > + * btree block is defined as:
> > > + *
> > > + * npb =  nr_items / blocks
> > > + *
> > > + * Some of the leftmost blocks in the level will contain one extra record as
> > > + * needed to handle uneven division.  If the number of records in any block
> > > + * would exceed maxrecs for that level, blocks is incremented and npb is
> > > + * recalculated.
> > > + *
> > > + * In other words, we compute the number of blocks needed to satisfy a given
> > > + * loading level, then spread the items as evenly as possible.
> > > + *
> > > + * To complete this step, call xfs_btree_bload_compute_geometry, which uses
> > > + * those settings to compute the height of the btree and the number of blocks
> > > + * that will be needed to construct the btree.  These values are stored in the
> > > + * @btree_height and @nr_blocks fields.
> > > + *
> > > + * At this point, the caller must allocate @nr_blocks blocks and save them for
> > > + * later.  If space is to be allocated transactionally, the staging cursor
> > > + * must be deleted before and recreated after, which is why computing the
> > > + * geometry is a separate step.
> > > + *
> > > + * The fourth step in the bulk loading process is to set the function pointers
> > > + * in the bload context structure.  @get_data will be called for each record
> > > + * that will be loaded into the btree; it should set the cursor's bc_rec
> > > + * field, which will be converted to on-disk format and copied into the
> > > + * appropriate record slot.  @alloc_block should supply one of the blocks
> > > + * allocated in the previous step.  For btrees which are rooted in an inode
> > > + * fork, @iroot_size is called to compute the size of the incore btree root
> > > + * block.  Call xfs_btree_bload to start constructing the btree.
> > > + *
> > > + * The final step is to commit the staging cursor, which logs the new btree
> > > + * root and turns the btree into a regular btree cursor, and free the fake
> > > + * roots.
> > > + */
> > > +
> > > +/*
> > > + * Put a btree block that we're loading onto the ordered list and release it.
> > > + * The btree blocks will be written when the final transaction swapping the
> > > + * btree roots is committed.
> > > + */
> > > +static void
> > > +xfs_btree_bload_drop_buf(
> > > +	struct xfs_trans	*tp,
> > > +	struct xfs_buf		**bpp)
> > > +{
> > > +	if (*bpp == NULL)
> > > +		return;
> > > +
> > > +	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_BTREE_BUF);
> > > +	xfs_trans_ordered_buf(tp, *bpp);
> > > +	xfs_trans_brelse(tp, *bpp);
> > > +	*bpp = NULL;
> > > +}
> > > +
> > > +/* Allocate and initialize one btree block for bulk loading. */
> > > +STATIC int
> > > +xfs_btree_bload_prep_block(
> > > +	struct xfs_btree_cur		*cur,
> > > +	struct xfs_btree_bload		*bbl,
> > > +	unsigned int			level,
> > > +	unsigned int			nr_this_block,
> > > +	union xfs_btree_ptr		*ptrp,
> > > +	struct xfs_buf			**bpp,
> > > +	struct xfs_btree_block		**blockp,
> > > +	void				*priv)
> > > +{
> > > +	union xfs_btree_ptr		new_ptr;
> > > +	struct xfs_buf			*new_bp;
> > > +	struct xfs_btree_block		*new_block;
> > > +	int				ret;
> > > +
> > > +	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
> > > +	    level == cur->bc_nlevels - 1) {
> > > +		struct xfs_ifork	*ifp = cur->bc_private.b.ifake->if_fork;
> > > +		size_t			new_size;
> > > +
> > > +		/* Allocate a new incore btree root block. */
> > > +		new_size = bbl->iroot_size(cur, nr_this_block, priv);
> > > +		ifp->if_broot = kmem_zalloc(new_size, 0);
> > > +		ifp->if_broot_bytes = (int)new_size;
> > > +		ifp->if_flags |= XFS_IFBROOT;
> > > +
> > > +		/* Initialize it and send it out. */
> > > +		xfs_btree_init_block_int(cur->bc_mp, ifp->if_broot,
> > > +				XFS_BUF_DADDR_NULL, cur->bc_btnum, level,
> > > +				nr_this_block, cur->bc_private.b.ip->i_ino,
> > > +				cur->bc_flags);
> > > +
> > > +		*bpp = NULL;
> > > +		*blockp = ifp->if_broot;
> > > +		xfs_btree_set_ptr_null(cur, ptrp);
> > > +		return 0;
> > > +	}
> > > +
> > > +	/* Allocate a new leaf block. */
> > > +	ret = bbl->alloc_block(cur, &new_ptr, priv);
> > > +	if (ret)
> > > +		return ret;
> > > +
> > > +	ASSERT(!xfs_btree_ptr_is_null(cur, &new_ptr));
> > > +
> > > +	ret = xfs_btree_get_buf_block(cur, &new_ptr, &new_block, &new_bp);
> > > +	if (ret)
> > > +		return ret;
> > > +
> > > +	/* Initialize the btree block. */
> > > +	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
> > > +	if (*blockp)
> > > +		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
> > > +	xfs_btree_set_sibling(cur, new_block, ptrp, XFS_BB_LEFTSIB);
> > > +	xfs_btree_set_numrecs(new_block, nr_this_block);
> > > +
> > > +	/* Release the old block and set the out parameters. */
> > > +	xfs_btree_bload_drop_buf(cur->bc_tp, bpp);
> > > +	*blockp = new_block;
> > > +	*bpp = new_bp;
> > > +	xfs_btree_copy_ptrs(cur, ptrp, &new_ptr, 1);
> > > +	return 0;
> > > +}
> > > +
> > > +/* Load one leaf block. */
> > > +STATIC int
> > > +xfs_btree_bload_leaf(
> > > +	struct xfs_btree_cur		*cur,
> > > +	unsigned int			recs_this_block,
> > > +	xfs_btree_bload_get_fn		get_data,
> > > +	struct xfs_btree_block		*block,
> > > +	void				*priv)
> > > +{
> > > +	unsigned int			j;
> > > +	int				ret;
> > > +
> > > +	/* Fill the leaf block with records. */
> > > +	for (j = 1; j <= recs_this_block; j++) {
> > > +		union xfs_btree_rec	*block_recs;
> > > +
> > > +		ret = get_data(cur, priv);
> > > +		if (ret)
> > > +			return ret;
> > > +		block_recs = xfs_btree_rec_addr(cur, j, block);
> > > +		cur->bc_ops->init_rec_from_cur(cur, block_recs);
> > > +	}
> > > +
> > > +	return 0;
> > > +}
> > > +
> > > +/* Load one node block. */
> > > +STATIC int
> > > +xfs_btree_bload_node(
> > > +	struct xfs_btree_cur	*cur,
> > > +	unsigned int		recs_this_block,
> > > +	union xfs_btree_ptr	*child_ptr,
> > > +	struct xfs_btree_block	*block)
> > > +{
> > > +	unsigned int		j;
> > > +	int			ret;
> > > +
> > > +	/* Fill the node block with keys and pointers. */
> > > +	for (j = 1; j <= recs_this_block; j++) {
> > > +		union xfs_btree_key	child_key;
> > > +		union xfs_btree_ptr	*block_ptr;
> > > +		union xfs_btree_key	*block_key;
> > > +		struct xfs_btree_block	*child_block;
> > > +		struct xfs_buf		*child_bp;
> > > +
> > > +		ASSERT(!xfs_btree_ptr_is_null(cur, child_ptr));
> > > +
> > > +		ret = xfs_btree_get_buf_block(cur, child_ptr, &child_block,
> > > +				&child_bp);
> > > +		if (ret)
> > > +			return ret;
> > > +
> > > +		xfs_btree_get_keys(cur, child_block, &child_key);
> > > +
> > > +		block_ptr = xfs_btree_ptr_addr(cur, j, block);
> > > +		xfs_btree_copy_ptrs(cur, block_ptr, child_ptr, 1);
> > > +
> > > +		block_key = xfs_btree_key_addr(cur, j, block);
> > > +		xfs_btree_copy_keys(cur, block_key, &child_key, 1);
> > > +
> > > +		xfs_btree_get_sibling(cur, child_block, child_ptr,
> > > +				XFS_BB_RIGHTSIB);
> > > +		xfs_trans_brelse(cur->bc_tp, child_bp);
> > > +	}
> > > +
> > > +	return 0;
> > > +}
> > > +
> > > +/*
> > > + * Compute the maximum number of records (or keyptrs) per block that we want to
> > > + * install at this level in the btree.  Caller is responsible for having set
> > > + * @cur->bc_private.b.forksize to the desired fork size, if appropriate.
> > > + */
> > > +STATIC unsigned int
> > > +xfs_btree_bload_max_npb(
> > > +	struct xfs_btree_cur	*cur,
> > > +	struct xfs_btree_bload	*bbl,
> > > +	unsigned int		level)
> > > +{
> > > +	unsigned int		ret;
> > > +
> > > +	if (level == cur->bc_nlevels - 1 && cur->bc_ops->get_dmaxrecs)
> > > +		return cur->bc_ops->get_dmaxrecs(cur, level);
> > > +
> > > +	ret = cur->bc_ops->get_maxrecs(cur, level);
> > > +	if (level == 0)
> > > +		ret -= bbl->leaf_slack;
> > > +	else
> > > +		ret -= bbl->node_slack;
> > > +	return ret;
> > > +}
> > > +
> > > +/*
> > > + * Compute the desired number of records (or keyptrs) per block that we want to
> > > + * install at this level in the btree, which must be somewhere between minrecs
> > > + * and max_npb.  The caller is free to install fewer records per block.
> > > + */
> > > +STATIC unsigned int
> > > +xfs_btree_bload_desired_npb(
> > > +	struct xfs_btree_cur	*cur,
> > > +	struct xfs_btree_bload	*bbl,
> > > +	unsigned int		level)
> > > +{
> > > +	unsigned int		npb = xfs_btree_bload_max_npb(cur, bbl, level);
> > > +
> > > +	/* Root blocks are not subject to minrecs rules. */
> > > +	if (level == cur->bc_nlevels - 1)
> > > +		return max(1U, npb);
> > > +
> > > +	return max_t(unsigned int, cur->bc_ops->get_minrecs(cur, level), npb);
> > > +}
> > > +
> > > +/*
> > > + * Compute the number of records to be stored in each block at this level and
> > > + * the number of blocks for this level.  For leaf levels, we must populate an
> > > + * empty root block even if there are no records, so we have to have at least
> > > + * one block.
> > > + */
> > > +STATIC void
> > > +xfs_btree_bload_level_geometry(
> > > +	struct xfs_btree_cur	*cur,
> > > +	struct xfs_btree_bload	*bbl,
> > > +	unsigned int		level,
> > > +	uint64_t		nr_this_level,
> > > +	unsigned int		*avg_per_block,
> > > +	uint64_t		*blocks,
> > > +	uint64_t		*blocks_with_extra)
> > > +{
> > > +	uint64_t		npb;
> > > +	uint64_t		dontcare;
> > > +	unsigned int		desired_npb;
> > > +	unsigned int		maxnr;
> > > +
> > > +	maxnr = cur->bc_ops->get_maxrecs(cur, level);
> > > +
> > > +	/*
> > > +	 * Compute the number of blocks we need to fill each block with the
> > > +	 * desired number of records/keyptrs per block.  Because desired_npb
> > > +	 * could be minrecs, we use regular integer division (which rounds
> > > +	 * the block count down) so that in the next step the effective # of
> > > +	 * items per block will never be less than desired_npb.
> > > +	 */
> > > +	desired_npb = xfs_btree_bload_desired_npb(cur, bbl, level);
> > > +	*blocks = div64_u64_rem(nr_this_level, desired_npb, &dontcare);
> > > +	*blocks = max(1ULL, *blocks);
> > > +
> > > +	/*
> > > +	 * Compute the number of records that we will actually put in each
> > > +	 * block, assuming that we want to spread the records evenly between
> > > +	 * the blocks.  Take care that the effective # of items per block (npb)
> > > +	 * won't exceed maxrecs even for the blocks that get an extra record,
> > > +	 * since desired_npb could be maxrecs, and in the previous step we
> > > +	 * rounded the block count down.
> > > +	 */
> > > +	npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > +	if (npb > maxnr || (npb == maxnr && *blocks_with_extra > 0)) {
> > > +		(*blocks)++;
> > > +		npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > +	}
> > > +
> > > +	*avg_per_block = min_t(uint64_t, npb, nr_this_level);
> > > +
> > > +	trace_xfs_btree_bload_level_geometry(cur, level, nr_this_level,
> > > +			*avg_per_block, desired_npb, *blocks,
> > > +			*blocks_with_extra);
> > > +}
> > > +
> > > +/*
> > > + * Ensure a slack value is appropriate for the btree.
> > > + *
> > > + * If the slack value is negative, set slack so that we fill the block to
> > > + * halfway between minrecs and maxrecs.  Make sure the slack is never so large
> > > + * that we can underflow minrecs.
> > > + */
> > > +static void
> > > +xfs_btree_bload_ensure_slack(
> > > +	struct xfs_btree_cur	*cur,
> > > +	int			*slack,
> > > +	int			level)
> > > +{
> > > +	int			maxr;
> > > +	int			minr;
> > > +
> > > +	/*
> > > +	 * We only care about slack for btree blocks, so set the btree nlevels
> > > +	 * to 3 so that level 0 is a leaf block and level 1 is a node block.
> > > +	 * Avoid straying into inode roots, since we don't do slack there.
> > > +	 */
> > > +	cur->bc_nlevels = 3;
> > > +	maxr = cur->bc_ops->get_maxrecs(cur, level);
> > > +	minr = cur->bc_ops->get_minrecs(cur, level);
> > > +
> > > +	/*
> > > +	 * If slack is negative, automatically set slack so that we load the
> > > +	 * btree block approximately halfway between minrecs and maxrecs.
> > > +	 * Generally, this will net us 75% loading.
> > > +	 */
> > > +	if (*slack < 0)
> > > +		*slack = maxr - ((maxr + minr) >> 1);
> > > +
> > > +	*slack = min(*slack, maxr - minr);
> > > +}
> > > +
> > > +/*
> > > + * Prepare a btree cursor for a bulk load operation by computing the geometry
> > > + * fields in @bbl.  Caller must ensure that the btree cursor is a staging
> > > + * cursor.  This function can be called multiple times.
> > > + */
> > > +int
> > > +xfs_btree_bload_compute_geometry(
> > > +	struct xfs_btree_cur	*cur,
> > > +	struct xfs_btree_bload	*bbl,
> > > +	uint64_t		nr_records)
> > > +{
> > > +	uint64_t		nr_blocks = 0;
> > > +	uint64_t		nr_this_level;
> > > +
> > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > +
> > > +	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
> > > +	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
> > > +
> > > +	bbl->nr_records = nr_this_level = nr_records;
> > > +	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
> > > +		uint64_t	level_blocks;
> > > +		uint64_t	dontcare64;
> > > +		unsigned int	level = cur->bc_nlevels - 1;
> > > +		unsigned int	avg_per_block;
> > > +
> > > +		/*
> > > +		 * If all the things we want to store at this level would fit
> > > +		 * in a single root block, then we have our btree root and are
> > > +		 * done.  Note that bmap btrees do not allow records in the
> > > +		 * root.
> > > +		 */
> > > +		if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) || level != 0) {
> > > +			xfs_btree_bload_level_geometry(cur, bbl, level,
> > > +					nr_this_level, &avg_per_block,
> > > +					&level_blocks, &dontcare64);
> > > +			if (nr_this_level <= avg_per_block) {
> > > +				nr_blocks++;
> > > +				break;
> > > +			}
> > > +		}
> > > +
> > > +		/*
> > > +		 * Otherwise, we have to store all the records for this level
> > > +		 * in blocks and therefore need another level of btree to point
> > > +		 * to those blocks.  Increase the number of levels and
> > > +		 * recompute the number of records we can store at this level
> > > +		 * because that can change depending on whether or not a level
> > > +		 * is the root level.
> > > +		 */
> > > +		cur->bc_nlevels++;
> > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > +				&avg_per_block, &level_blocks, &dontcare64);
> > > +		nr_blocks += level_blocks;
> > > +		nr_this_level = level_blocks;
> > > +	}
> > > +
> > > +	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
> > > +		return -EOVERFLOW;
> > > +
> > > +	bbl->btree_height = cur->bc_nlevels;
> > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
> > > +		bbl->nr_blocks = nr_blocks - 1;
> > > +	else
> > > +		bbl->nr_blocks = nr_blocks;
> > > +	return 0;
> > > +}
> > > +
> > > +/*
> > > + * Bulk load a btree.
> > > + *
> > > + * Load @bbl->nr_records quantity of records into a btree using the supplied
> > > + * empty and staging btree cursor @cur and a @bbl that has been filled out by
> > > + * the xfs_btree_bload_compute_geometry function.
> > > + *
> > > + * The @bbl->get_data function must populate the cursor's bc_rec every time it
> > > + * is called.  The @bbl->alloc_block function will be used to allocate new
> > > + * btree blocks.  @priv is passed to both functions.
> > > + *
> > > + * Caller must ensure that @cur is a staging cursor.  Any existing btree rooted
> > > + * in the fakeroot will be lost, so do not call this function twice.
> > > + */
> > > +int
> > > +xfs_btree_bload(
> > > +	struct xfs_btree_cur		*cur,
> > > +	struct xfs_btree_bload		*bbl,
> > > +	void				*priv)
> > > +{
> > > +	union xfs_btree_ptr		child_ptr;
> > > +	union xfs_btree_ptr		ptr;
> > > +	struct xfs_buf			*bp = NULL;
> > > +	struct xfs_btree_block		*block = NULL;
> > > +	uint64_t			nr_this_level = bbl->nr_records;
> > > +	uint64_t			blocks;
> > > +	uint64_t			i;
> > > +	uint64_t			blocks_with_extra;
> > > +	uint64_t			total_blocks = 0;
> > > +	unsigned int			avg_per_block;
> > > +	unsigned int			level = 0;
> > > +	int				ret;
> > > +
> > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > +
> > > +	cur->bc_nlevels = bbl->btree_height;
> > > +	xfs_btree_set_ptr_null(cur, &child_ptr);
> > > +	xfs_btree_set_ptr_null(cur, &ptr);
> > > +
> > > +	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > +			&avg_per_block, &blocks, &blocks_with_extra);
> > > +
> > > +	/* Load each leaf block. */
> > > +	for (i = 0; i < blocks; i++) {
> > > +		unsigned int		nr_this_block = avg_per_block;
> > > +
> > > +		if (i < blocks_with_extra)
> > > +			nr_this_block++;
> > > +
> > > +		ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > +				nr_this_block, &ptr, &bp, &block, priv);
> > > +		if (ret)
> > > +			return ret;
> > > +
> > > +		trace_xfs_btree_bload_block(cur, level, i, blocks, &ptr,
> > > +				nr_this_block);
> > > +
> > > +		ret = xfs_btree_bload_leaf(cur, nr_this_block, bbl->get_data,
> > > +				block, priv);
> > > +		if (ret)
> > > +			goto out;
> > > +
> > > +		/* Record the leftmost pointer to start the next level. */
> > > +		if (i == 0)
> > > +			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
> > > +	}
> > > +	total_blocks += blocks;
> > > +	xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > +
> > > +	/* Populate the internal btree nodes. */
> > > +	for (level = 1; level < cur->bc_nlevels; level++) {
> > > +		union xfs_btree_ptr	first_ptr;
> > > +
> > > +		nr_this_level = blocks;
> > > +		block = NULL;
> > > +		xfs_btree_set_ptr_null(cur, &ptr);
> > > +
> > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > +				&avg_per_block, &blocks, &blocks_with_extra);
> > > +
> > > +		/* Load each node block. */
> > > +		for (i = 0; i < blocks; i++) {
> > > +			unsigned int	nr_this_block = avg_per_block;
> > > +
> > > +			if (i < blocks_with_extra)
> > > +				nr_this_block++;
> > > +
> > > +			ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > +					nr_this_block, &ptr, &bp, &block,
> > > +					priv);
> > > +			if (ret)
> > > +				return ret;
> > > +
> > > +			trace_xfs_btree_bload_block(cur, level, i, blocks,
> > > +					&ptr, nr_this_block);
> > > +
> > > +			ret = xfs_btree_bload_node(cur, nr_this_block,
> > > +					&child_ptr, block);
> > > +			if (ret)
> > > +				goto out;
> > > +
> > > +			/*
> > > +			 * Record the leftmost pointer to start the next level.
> > > +			 */
> > > +			if (i == 0)
> > > +				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
> > > +		}
> > > +		total_blocks += blocks;
> > > +		xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > +		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
> > > +	}
> > > +
> > > +	/* Initialize the new root. */
> > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
> > > +		ASSERT(xfs_btree_ptr_is_null(cur, &ptr));
> > > +		cur->bc_private.b.ifake->if_levels = cur->bc_nlevels;
> > > +		cur->bc_private.b.ifake->if_blocks = total_blocks - 1;
> > > +	} else {
> > > +		cur->bc_private.a.afake->af_root = be32_to_cpu(ptr.s);
> > > +		cur->bc_private.a.afake->af_levels = cur->bc_nlevels;
> > > +		cur->bc_private.a.afake->af_blocks = total_blocks;
> > > +	}
> > > +out:
> > > +	if (bp)
> > > +		xfs_trans_brelse(cur->bc_tp, bp);
> > > +	return ret;
> > > +}
> > > diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
> > > index a17becb72ab8..5c6992a04ea2 100644
> > > --- a/fs/xfs/libxfs/xfs_btree.h
> > > +++ b/fs/xfs/libxfs/xfs_btree.h
> > > @@ -582,4 +582,47 @@ void xfs_btree_stage_ifakeroot(struct xfs_btree_cur *cur,
> > >  void xfs_btree_commit_ifakeroot(struct xfs_btree_cur *cur, int whichfork,
> > >  		const struct xfs_btree_ops *ops);
> > >  
> > > +typedef int (*xfs_btree_bload_get_fn)(struct xfs_btree_cur *cur, void *priv);
> > > +typedef int (*xfs_btree_bload_alloc_block_fn)(struct xfs_btree_cur *cur,
> > > +		union xfs_btree_ptr *ptr, void *priv);
> > > +typedef size_t (*xfs_btree_bload_iroot_size_fn)(struct xfs_btree_cur *cur,
> > > +		unsigned int nr_this_level, void *priv);
> > > +
> > > +/* Bulk loading of staged btrees. */
> > > +struct xfs_btree_bload {
> > > +	/* Function to store a record in the cursor. */
> > > +	xfs_btree_bload_get_fn		get_data;
> > > +
> > > +	/* Function to allocate a block for the btree. */
> > > +	xfs_btree_bload_alloc_block_fn	alloc_block;
> > > +
> > > +	/* Function to compute the size of the in-core btree root block. */
> > > +	xfs_btree_bload_iroot_size_fn	iroot_size;
> > > +
> > > +	/* Number of records the caller wants to store. */
> > > +	uint64_t			nr_records;
> > > +
> > > +	/* Number of btree blocks needed to store those records. */
> > > +	uint64_t			nr_blocks;
> > > +
> > > +	/*
> > > +	 * Number of free records to leave in each leaf block.  If this (or
> > > +	 * any of the slack values) are negative, this will be computed to
> > > +	 * be halfway between maxrecs and minrecs.  This typically leaves the
> > > +	 * block 75% full.
> > > +	 */
> > > +	int				leaf_slack;
> > > +
> > > +	/* Number of free keyptrs to leave in each node block. */
> > > +	int				node_slack;
> > > +
> > > +	/* Computed btree height. */
> > > +	unsigned int			btree_height;
> > > +};
> > > +
> > > +int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
> > > +		struct xfs_btree_bload *bbl, uint64_t nr_records);
> > > +int xfs_btree_bload(struct xfs_btree_cur *cur, struct xfs_btree_bload *bbl,
> > > +		void *priv);
> > > +
> > >  #endif	/* __XFS_BTREE_H__ */
> > > diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
> > > index bc85b89f88ca..9b5e58a92381 100644
> > > --- a/fs/xfs/xfs_trace.c
> > > +++ b/fs/xfs/xfs_trace.c
> > > @@ -6,6 +6,7 @@
> > >  #include "xfs.h"
> > >  #include "xfs_fs.h"
> > >  #include "xfs_shared.h"
> > > +#include "xfs_bit.h"
> > >  #include "xfs_format.h"
> > >  #include "xfs_log_format.h"
> > >  #include "xfs_trans_resv.h"
> > > diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
> > > index a78055521fcd..6d7ba64b7a0f 100644
> > > --- a/fs/xfs/xfs_trace.h
> > > +++ b/fs/xfs/xfs_trace.h
> > > @@ -35,6 +35,7 @@ struct xfs_icreate_log;
> > >  struct xfs_owner_info;
> > >  struct xfs_trans_res;
> > >  struct xfs_inobt_rec_incore;
> > > +union xfs_btree_ptr;
> > >  
> > >  DECLARE_EVENT_CLASS(xfs_attr_list_class,
> > >  	TP_PROTO(struct xfs_attr_list_context *ctx),
> > > @@ -3670,6 +3671,90 @@ TRACE_EVENT(xfs_btree_commit_ifakeroot,
> > >  		  __entry->blocks)
> > >  )
> > >  
> > > +TRACE_EVENT(xfs_btree_bload_level_geometry,
> > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > +		 uint64_t nr_this_level, unsigned int nr_per_block,
> > > +		 unsigned int desired_npb, uint64_t blocks,
> > > +		 uint64_t blocks_with_extra),
> > > +	TP_ARGS(cur, level, nr_this_level, nr_per_block, desired_npb, blocks,
> > > +		blocks_with_extra),
> > > +	TP_STRUCT__entry(
> > > +		__field(dev_t, dev)
> > > +		__field(xfs_btnum_t, btnum)
> > > +		__field(unsigned int, level)
> > > +		__field(unsigned int, nlevels)
> > > +		__field(uint64_t, nr_this_level)
> > > +		__field(unsigned int, nr_per_block)
> > > +		__field(unsigned int, desired_npb)
> > > +		__field(unsigned long long, blocks)
> > > +		__field(unsigned long long, blocks_with_extra)
> > > +	),
> > > +	TP_fast_assign(
> > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > +		__entry->btnum = cur->bc_btnum;
> > > +		__entry->level = level;
> > > +		__entry->nlevels = cur->bc_nlevels;
> > > +		__entry->nr_this_level = nr_this_level;
> > > +		__entry->nr_per_block = nr_per_block;
> > > +		__entry->desired_npb = desired_npb;
> > > +		__entry->blocks = blocks;
> > > +		__entry->blocks_with_extra = blocks_with_extra;
> > > +	),
> > > +	TP_printk("dev %d:%d btree %s level %u/%u nr_this_level %llu nr_per_block %u desired_npb %u blocks %llu blocks_with_extra %llu",
> > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > +		  __entry->level,
> > > +		  __entry->nlevels,
> > > +		  __entry->nr_this_level,
> > > +		  __entry->nr_per_block,
> > > +		  __entry->desired_npb,
> > > +		  __entry->blocks,
> > > +		  __entry->blocks_with_extra)
> > > +)
> > > +
> > > +TRACE_EVENT(xfs_btree_bload_block,
> > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > +		 uint64_t block_idx, uint64_t nr_blocks,
> > > +		 union xfs_btree_ptr *ptr, unsigned int nr_records),
> > > +	TP_ARGS(cur, level, block_idx, nr_blocks, ptr, nr_records),
> > > +	TP_STRUCT__entry(
> > > +		__field(dev_t, dev)
> > > +		__field(xfs_btnum_t, btnum)
> > > +		__field(unsigned int, level)
> > > +		__field(unsigned long long, block_idx)
> > > +		__field(unsigned long long, nr_blocks)
> > > +		__field(xfs_agnumber_t, agno)
> > > +		__field(xfs_agblock_t, agbno)
> > > +		__field(unsigned int, nr_records)
> > > +	),
> > > +	TP_fast_assign(
> > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > +		__entry->btnum = cur->bc_btnum;
> > > +		__entry->level = level;
> > > +		__entry->block_idx = block_idx;
> > > +		__entry->nr_blocks = nr_blocks;
> > > +		if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
> > > +			xfs_fsblock_t	fsb = be64_to_cpu(ptr->l);
> > > +
> > > +			__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsb);
> > > +			__entry->agbno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsb);
> > > +		} else {
> > > +			__entry->agno = cur->bc_private.a.agno;
> > > +			__entry->agbno = be32_to_cpu(ptr->s);
> > > +		}
> > > +		__entry->nr_records = nr_records;
> > > +	),
> > > +	TP_printk("dev %d:%d btree %s level %u block %llu/%llu fsb (%u/%u) recs %u",
> > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > +		  __entry->level,
> > > +		  __entry->block_idx,
> > > +		  __entry->nr_blocks,
> > > +		  __entry->agno,
> > > +		  __entry->agbno,
> > > +		  __entry->nr_records)
> > > +)
> > > +
> > >  #endif /* _TRACE_XFS_H */
> > >  
> > >  #undef TRACE_INCLUDE_PATH
> > >

Darrick J. Wong Oct. 17, 2019, 12:40 a.m. UTC | #4

On Wed, Oct 16, 2019 at 05:07:31PM -0400, Brian Foster wrote:
> On Wed, Oct 16, 2019 at 11:15:02AM -0700, Darrick J. Wong wrote:
> > On Wed, Oct 16, 2019 at 11:26:48AM -0400, Brian Foster wrote:
> > > On Wed, Oct 09, 2019 at 09:48:18AM -0700, Darrick J. Wong wrote:
> > > > From: Darrick J. Wong <[e-mail]>
> > > > 
> > > > Add a new btree function that enables us to bulk load a btree cursor.
> > > > This will be used by the upcoming online repair patches to generate new
> > > > btrees.
> > > > 
> > > > Signed-off-by: Darrick J. Wong <[e-mail]>
> > > > ---
> > > >  fs/xfs/libxfs/xfs_btree.c |  566 +++++++++++++++++++++++++++++++++++++++++++++
> > > >  fs/xfs/libxfs/xfs_btree.h |   43 +++
> > > >  fs/xfs/xfs_trace.c        |    1 
> > > >  fs/xfs/xfs_trace.h        |   85 +++++++
> > > >  4 files changed, 694 insertions(+), 1 deletion(-)
> > > > 
> > > > 
> > > > diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
> > > > index 4b06d5d86834..17b0fdb87729 100644
> > > > --- a/fs/xfs/libxfs/xfs_btree.c
> > > > +++ b/fs/xfs/libxfs/xfs_btree.c
> > > ...
> > > > @@ -5104,3 +5104,567 @@ xfs_btree_commit_ifakeroot(
> > > >  	cur->bc_ops = ops;
> > > >  	cur->bc_flags &= ~XFS_BTREE_STAGING;
> > > >  }
> > > > +
> > > > +/*
> > > > + * Bulk Loading of Staged Btrees
> > > > + * =============================
> > > > + *
> > > > + * This interface is used with a staged btree cursor to create a totally new
> > > > + * btree with a large number of records (i.e. more than what would fit in a
> > > > + * single block).  When the creation is complete, the new root can be linked
> > > > + * atomically into the filesystem by committing the staged cursor.
> > > > + *
> > 
> > [paraphrasing a conversation we had on irc]
> > 
> > > Thanks for the documentation. So what is the purpose behind the whole
> > > bulk loading thing as opposed to something like faking up an AG
> > > structure (i.e. AGF) somewhere and using the existing cursor mechanisms
> > > (or something closer to it) to copy records from one place to another?
> > > Is it purely a performance/efficiency tradeoff? Bulk block allocation
> > > issues? Transactional/atomicity issues? All (or none :P) of the above?
> > 
> > Prior to the v20, the online repair series created a new btree root,
> > committed that into wherever the root lived, and inserted records one by
> > one into the btree.  There were quite a few drawbacks to this method:
> > 
> > 1. Inserting records one at a time can involve walking up the tree to
> > update node block pointers, which isn't terribly efficient if we're
> > likely going to rewrite the pointers (and relogging nodes) several more
> > times.
> > 
> > 2. Inserting records one at a time tends to leave a lot of half-empty
> > btree blocks because when one block fills up we split it and push half
> > the records to the new block.  It would be nice not to explode the size
> > of the btrees, and it would be particularly useful if we could control
> > the load factor of the new btree precisely.
> > 
> 
> Interesting... this is a trait the traditional btree update paths share
> though, right?

Right.  It's similar to the behavior Dave was seeing a couple of weeks
ago with Zorro's stress testing of the incore extent cache.

> > 3. The rebuild wasn't atomic, since we were replacing the root prior to
> > the insert loop.  If we crashed midway through a rebuild we'd end up
> > with a garbage btree and no indication that it was incorrect.  That's
> > how the fakeroot code got started.
> > 
> 
> Indeed, though this seems more related to the anchoring (i.e. fake root)
> approach than bulk vs. iterative construction.

Correct.

> > 4. In a previous version of the repair series I tried to batch as many
> > insert operations into a single transaction as possible, but my
> > transaction reservation fullness estimation function didn't work
> > reliably (particularly when things got really fragmented), so I backed
> > off to rolling after /every/ insertion.  That works well enough, but at
> > a cost of a lot of transaction rolling, which means that repairs plod
> > along very slowly.
> > 
> > 5. Performing an insert loop means that the btree blocks are allocated
> > one at a time as the btree expands.  This is suboptimal since we can
> > calculate the exact size of the new btree prior to building it, which
> > gives us the opportunity to recreate the index in a set of contiguous
> > blocks instead of scattering them.
> > 
> 
> Yep, FWIW it sounds like most of these tradeoffs are around
> performance/efficiency. 

<nod>

> > 6. If we crash midway through a rebuild, XFS neither cleaned up the mess
> > nor informed the administrator that it was necessary to re-run xfs_scrub
> > or xfs_repair to clean up the lost blocks.  Obviously, automatic cleanup
> > is a far better solution.
> > 
> 
> Similar to above, I think this kind of depends more on how/where to
> anchor an in-progress tree as opposed to what level records are copied
> at.

<nod> The six points are indeed the overall list of complaints about the
v19 code. :)

> > The first thing I decided to solve was the lack of atomicity.
> > 
> > For AG-rooted btrees, I thought about creating a fake xfs_buf for an AG
> > header buffer and extracting the root/level values after construction
> > completes.  That's possible, but it's risky because the fake buffer
> > could get logged and if the sector number matches the actual header
> > then it introduces buffer cache aliasing issues.
> > 
> > For inode-rooted btrees, one could create a fake xfs_inode with the same
> > i_ino as the target.  That presents the same aliasing issues as the fake
> > xfs_buf above.  A different strategy would be to allocate an unlinked
> > inode and then use the bmbt owner change (a.k.a. extent swap) to move
> > the mappings over.  That would work, though it has two large drawbacks:
> > (a) a lot of additional complexity around allocating and freeing the
> > temporary inode; and (b) future inode-rooted btrees such as the realtime
> > rmap btree would also have to implement an owner-change operation.
> > 
> 
> I was wondering more along the lines of having an actual anchor
> somewhere. E.g., think of it as a temporary/inaccessible location of a
> legitimate on-disk structure as opposed to a fake object in memory
> somewhere. A hidden/internal repair inode or some such, perhaps. I'm
> sure there's new code/complexity that would come around with that, but I
> think that's going to be unavoidable to some degree for an online repair
> mechanism. ;)

<nod> So far I /think/ I've managed to keep to an absolute minimum the
amount of metadata that gets written to disk prior to the commit.  I
haven't reread the series with an eye for how v20 is going to come up
short though. :)

I may very well have to revisit the hidden/internal repair inode concept
whenever I start working on rebuilding directories and xattrs since I
can't see any other way of atomically rebuilding those.  But that's
very very far out still.

> Note that this is all just handwaving on my part and still without full
> context as to how things are currently anchored, made atomic, etc. I'm
> primarily trying to understand the design reasoning based on the high
> level description.

<nod>

> > To fix (3), I thought it wise to have explicit fakeroot structures to
> > maintain a clean separation between what we're building and the rest of
> > the filesystem.  This also means that there's nothing on disk to clean
> > up if we fail at any point before we're ready to commit the new btree.
> > 
> 
> Hmm.. so this approach facilites a tree reconstruction in a single open
> transaction? If so, I suppose I could see some functional advantages to
> that.

Correct.

> > Then Dave (I think?) suggested that I  use EFIs strategically to
> > schedule freeing of the new btree blocks (the root commit transaction
> > would log EFDs to cancel them) and to schedule freeing of the old
> > blocks.  That solves (6), though the EFI wrangling doesn't happen for
> > another couple of series after this one.
> > 
> 
> Hm, Ok... so new btree block allocation(s?) in the same transaction as
> an EFI, to be processed on recovery if we crash, otherwise cancelled
> with an EFD on construction completion..?

Correct.  In the end, the transaction sequence looks like:

T[1]: Allocate an extent, log metadata updates to reflect that, log EFI
for the extent.

<repeat until we've allocated as many blocks as we need>

T[N]: Attach ordered buffers for the new btree's blocks.  Log the root
change.  Log EFDs for all the EFIs logged in T[1..N-1].  Log EFIs for
all the old btree blocks that we could find.

<roll transaction to write the ordered buffers and commit>

T[N+1]: Free an extent to finish the first EFI logged in the previous step.

<repeat until we've processed everything from the second wave of EFIs>

Call xfs_trans_commit and we're done.

> > He also suggested using ordered buffers to write out the new btree
> > blocks along with whatever logging was necessary to commit the new
> > btree.  It then occurred to me that xfs_repair open-codes the process of
> > calculating the geometry of a new btree, allocating all the blocks at
> > once, and writing out full btree blocks.  Somewhat annoyingly, it
> > features nearly the same (open-)code for all four AG btree types, which
> > is less maintainable than it could be.
> > 
> > I read through all four versions and used it to write the generic btree
> > bulk loading code.  For scrub I hooked that up to the "staged btree with
> > a fake root" stuff I'd already written, which solves (1), (2), (4), and
> > (5).
> > 
> > For xfsprogs[1], I deleted a few thousand lines of code from xfs_repair.
> > True, we don't reuse existing common code, but we at least get to share
> > new common btree code.
> > 
> 
> Yeah, the xfsprogs work certainly makes sense. Part of the reason I ask
> about this is the tradeoff of having multiple avenues to construct a
> tree in the kernel codebase.

<nod>

> > > This is my first pass through this so I'm mostly looking at big picture
> > > until I get to a point to see how these bits are used. The mechanism
> > > itself seems reasonable in principle, but the reason I ask is it also
> > > seems like there's inherent value in using more of same infrastructure
> > > to reconstruct a tree that we use to create one in the first place. We
> > > also already have primitives for things like fork swapping via the
> > > extent swap mechanism, etc.
> > 
> > "bfoster: I guess it would be nice to see that kind of make it work ->
> > make it fast evolution in tree"
> > 
> > For a while I did maintain the introduction of the bulk loading code as
> > separate patches against the v19 repair code, but unfortunately I
> > smushed them down before sending v20 to reduce the patch count, and
> > because I didn't want to argue with everyone over the semi-working code
> > that would then be replaced in the very next patch.
> > 
> 
> That's not quite what I meant... The approach you've taken makes sense
> to me for an implementation presented in a single series. I was more
> thinking that at the point where it was determined the implementation
> was going to change so drastically, after so many iterations it might
> have been useful to see the v19 approach merged in an experimental form

I would have liked to see the online repair stuff merged in experimental
form too so I can reduce the size of my patch queue, but oh well. :)

The silver lining to these lengthy reworks and slow review is that I can
come back and do a fresh self-review after a month and straighten out
the ugly parts as time goes by.  Unfortunately, that doesn't leave much
of a paper trail or obvious evidence of development history.

Eight months ago it occurred to me that perhaps there is some value in
retaining *some* periodic development history of this, so I've been
adding dated tags to my integration repo[1] based on my development
branch names, so I guess you could actually clone the git repo and git
diff from one tag to another.  In general I'll generate a new pile of
tags just before patchbombing.

(Granted 'repair-part-one' has been split into smaller parts now...)

> and then reworked upstream from there. Now that the new approach is
> implemented, I agree it's probably not worth reinserting the old
> approach at this point just to switch it out.

<nod>

> Thanks for the breakdown...

No problem, thanks for reading! :)

--D

[1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux.git

> Brian
> 
> > I could split them back out, though at a cost of having to reintroduce a
> > lot of hairy code in the bnobt/cntbt rebuild function to seed the free
> > new space btree root in order to make sure that the btree block
> > allocation code works properly, along with auditing the allocation paths
> > to make sure they don't use the old AGF or encounter other subtleties.
> > 
> > It'd be a lot of work considering that the v20 reconstruction code is
> > /much/ simpler than v19's was.  I also restructured the repair functions
> > to allocate one large context structure at the beginning instead of the
> > piecemeal way it was done onstack in v19 because stack usage was growing
> > close to 1k in some cases.
> > 
> > --D
> > 
> > [1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/log/?h=repair-bulk-load
> > 
> > > 
> > > Brian
> > > 
> > > > + * The first step for the caller is to construct a fake btree root structure
> > > > + * and a staged btree cursor.  A staging cursor contains all the geometry
> > > > + * information for the btree type but will fail all operations that could have
> > > > + * side effects in the filesystem (e.g. btree shape changes).  Regular
> > > > + * operations will not work unless the staging cursor is committed and becomes
> > > > + * a regular cursor.
> > > > + *
> > > > + * For a btree rooted in an AG header, use an xbtree_afakeroot structure.
> > > > + * This should be initialized to zero.  For a btree rooted in an inode fork,
> > > > + * use an xbtree_ifakeroot structure.  @if_fork_size field should be set to
> > > > + * the number of bytes available to the fork in the inode; @if_fork should
> > > > + * point to a freshly allocated xfs_inode_fork; and @if_format should be set
> > > > + * to the appropriate fork type (e.g. XFS_DINODE_FMT_BTREE).
> > > > + *
> > > > + * The next step for the caller is to initialize a struct xfs_btree_bload
> > > > + * context.  The @nr_records field is the number of records that are to be
> > > > + * loaded into the btree.  The @leaf_slack and @node_slack fields are the
> > > > + * number of records (or key/ptr) slots to leave empty in new btree blocks.
> > > > + * If a caller sets a slack value to -1, the slack value will be computed to
> > > > + * fill the block halfway between minrecs and maxrecs items per block.
> > > > + *
> > > > + * The number of items placed in each btree block is computed via the following
> > > > + * algorithm: For leaf levels, the number of items for the level is nr_records.
> > > > + * For node levels, the number of items for the level is the number of blocks
> > > > + * in the next lower level of the tree.  For each level, the desired number of
> > > > + * items per block is defined as:
> > > > + *
> > > > + * desired = max(minrecs, maxrecs - slack factor)
> > > > + *
> > > > + * The number of blocks for the level is defined to be:
> > > > + *
> > > > + * blocks = nr_items / desired
> > > > + *
> > > > + * Note this is rounded down so that the npb calculation below will never fall
> > > > + * below minrecs.  The number of items that will actually be loaded into each
> > > > + * btree block is defined as:
> > > > + *
> > > > + * npb =  nr_items / blocks
> > > > + *
> > > > + * Some of the leftmost blocks in the level will contain one extra record as
> > > > + * needed to handle uneven division.  If the number of records in any block
> > > > + * would exceed maxrecs for that level, blocks is incremented and npb is
> > > > + * recalculated.
> > > > + *
> > > > + * In other words, we compute the number of blocks needed to satisfy a given
> > > > + * loading level, then spread the items as evenly as possible.
> > > > + *
> > > > + * To complete this step, call xfs_btree_bload_compute_geometry, which uses
> > > > + * those settings to compute the height of the btree and the number of blocks
> > > > + * that will be needed to construct the btree.  These values are stored in the
> > > > + * @btree_height and @nr_blocks fields.
> > > > + *
> > > > + * At this point, the caller must allocate @nr_blocks blocks and save them for
> > > > + * later.  If space is to be allocated transactionally, the staging cursor
> > > > + * must be deleted before and recreated after, which is why computing the
> > > > + * geometry is a separate step.
> > > > + *
> > > > + * The fourth step in the bulk loading process is to set the function pointers
> > > > + * in the bload context structure.  @get_data will be called for each record
> > > > + * that will be loaded into the btree; it should set the cursor's bc_rec
> > > > + * field, which will be converted to on-disk format and copied into the
> > > > + * appropriate record slot.  @alloc_block should supply one of the blocks
> > > > + * allocated in the previous step.  For btrees which are rooted in an inode
> > > > + * fork, @iroot_size is called to compute the size of the incore btree root
> > > > + * block.  Call xfs_btree_bload to start constructing the btree.
> > > > + *
> > > > + * The final step is to commit the staging cursor, which logs the new btree
> > > > + * root and turns the btree into a regular btree cursor, and free the fake
> > > > + * roots.
> > > > + */
> > > > +
> > > > +/*
> > > > + * Put a btree block that we're loading onto the ordered list and release it.
> > > > + * The btree blocks will be written when the final transaction swapping the
> > > > + * btree roots is committed.
> > > > + */
> > > > +static void
> > > > +xfs_btree_bload_drop_buf(
> > > > +	struct xfs_trans	*tp,
> > > > +	struct xfs_buf		**bpp)
> > > > +{
> > > > +	if (*bpp == NULL)
> > > > +		return;
> > > > +
> > > > +	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_BTREE_BUF);
> > > > +	xfs_trans_ordered_buf(tp, *bpp);
> > > > +	xfs_trans_brelse(tp, *bpp);
> > > > +	*bpp = NULL;
> > > > +}
> > > > +
> > > > +/* Allocate and initialize one btree block for bulk loading. */
> > > > +STATIC int
> > > > +xfs_btree_bload_prep_block(
> > > > +	struct xfs_btree_cur		*cur,
> > > > +	struct xfs_btree_bload		*bbl,
> > > > +	unsigned int			level,
> > > > +	unsigned int			nr_this_block,
> > > > +	union xfs_btree_ptr		*ptrp,
> > > > +	struct xfs_buf			**bpp,
> > > > +	struct xfs_btree_block		**blockp,
> > > > +	void				*priv)
> > > > +{
> > > > +	union xfs_btree_ptr		new_ptr;
> > > > +	struct xfs_buf			*new_bp;
> > > > +	struct xfs_btree_block		*new_block;
> > > > +	int				ret;
> > > > +
> > > > +	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
> > > > +	    level == cur->bc_nlevels - 1) {
> > > > +		struct xfs_ifork	*ifp = cur->bc_private.b.ifake->if_fork;
> > > > +		size_t			new_size;
> > > > +
> > > > +		/* Allocate a new incore btree root block. */
> > > > +		new_size = bbl->iroot_size(cur, nr_this_block, priv);
> > > > +		ifp->if_broot = kmem_zalloc(new_size, 0);
> > > > +		ifp->if_broot_bytes = (int)new_size;
> > > > +		ifp->if_flags |= XFS_IFBROOT;
> > > > +
> > > > +		/* Initialize it and send it out. */
> > > > +		xfs_btree_init_block_int(cur->bc_mp, ifp->if_broot,
> > > > +				XFS_BUF_DADDR_NULL, cur->bc_btnum, level,
> > > > +				nr_this_block, cur->bc_private.b.ip->i_ino,
> > > > +				cur->bc_flags);
> > > > +
> > > > +		*bpp = NULL;
> > > > +		*blockp = ifp->if_broot;
> > > > +		xfs_btree_set_ptr_null(cur, ptrp);
> > > > +		return 0;
> > > > +	}
> > > > +
> > > > +	/* Allocate a new leaf block. */
> > > > +	ret = bbl->alloc_block(cur, &new_ptr, priv);
> > > > +	if (ret)
> > > > +		return ret;
> > > > +
> > > > +	ASSERT(!xfs_btree_ptr_is_null(cur, &new_ptr));
> > > > +
> > > > +	ret = xfs_btree_get_buf_block(cur, &new_ptr, &new_block, &new_bp);
> > > > +	if (ret)
> > > > +		return ret;
> > > > +
> > > > +	/* Initialize the btree block. */
> > > > +	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
> > > > +	if (*blockp)
> > > > +		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
> > > > +	xfs_btree_set_sibling(cur, new_block, ptrp, XFS_BB_LEFTSIB);
> > > > +	xfs_btree_set_numrecs(new_block, nr_this_block);
> > > > +
> > > > +	/* Release the old block and set the out parameters. */
> > > > +	xfs_btree_bload_drop_buf(cur->bc_tp, bpp);
> > > > +	*blockp = new_block;
> > > > +	*bpp = new_bp;
> > > > +	xfs_btree_copy_ptrs(cur, ptrp, &new_ptr, 1);
> > > > +	return 0;
> > > > +}
> > > > +
> > > > +/* Load one leaf block. */
> > > > +STATIC int
> > > > +xfs_btree_bload_leaf(
> > > > +	struct xfs_btree_cur		*cur,
> > > > +	unsigned int			recs_this_block,
> > > > +	xfs_btree_bload_get_fn		get_data,
> > > > +	struct xfs_btree_block		*block,
> > > > +	void				*priv)
> > > > +{
> > > > +	unsigned int			j;
> > > > +	int				ret;
> > > > +
> > > > +	/* Fill the leaf block with records. */
> > > > +	for (j = 1; j <= recs_this_block; j++) {
> > > > +		union xfs_btree_rec	*block_recs;
> > > > +
> > > > +		ret = get_data(cur, priv);
> > > > +		if (ret)
> > > > +			return ret;
> > > > +		block_recs = xfs_btree_rec_addr(cur, j, block);
> > > > +		cur->bc_ops->init_rec_from_cur(cur, block_recs);
> > > > +	}
> > > > +
> > > > +	return 0;
> > > > +}
> > > > +
> > > > +/* Load one node block. */
> > > > +STATIC int
> > > > +xfs_btree_bload_node(
> > > > +	struct xfs_btree_cur	*cur,
> > > > +	unsigned int		recs_this_block,
> > > > +	union xfs_btree_ptr	*child_ptr,
> > > > +	struct xfs_btree_block	*block)
> > > > +{
> > > > +	unsigned int		j;
> > > > +	int			ret;
> > > > +
> > > > +	/* Fill the node block with keys and pointers. */
> > > > +	for (j = 1; j <= recs_this_block; j++) {
> > > > +		union xfs_btree_key	child_key;
> > > > +		union xfs_btree_ptr	*block_ptr;
> > > > +		union xfs_btree_key	*block_key;
> > > > +		struct xfs_btree_block	*child_block;
> > > > +		struct xfs_buf		*child_bp;
> > > > +
> > > > +		ASSERT(!xfs_btree_ptr_is_null(cur, child_ptr));
> > > > +
> > > > +		ret = xfs_btree_get_buf_block(cur, child_ptr, &child_block,
> > > > +				&child_bp);
> > > > +		if (ret)
> > > > +			return ret;
> > > > +
> > > > +		xfs_btree_get_keys(cur, child_block, &child_key);
> > > > +
> > > > +		block_ptr = xfs_btree_ptr_addr(cur, j, block);
> > > > +		xfs_btree_copy_ptrs(cur, block_ptr, child_ptr, 1);
> > > > +
> > > > +		block_key = xfs_btree_key_addr(cur, j, block);
> > > > +		xfs_btree_copy_keys(cur, block_key, &child_key, 1);
> > > > +
> > > > +		xfs_btree_get_sibling(cur, child_block, child_ptr,
> > > > +				XFS_BB_RIGHTSIB);
> > > > +		xfs_trans_brelse(cur->bc_tp, child_bp);
> > > > +	}
> > > > +
> > > > +	return 0;
> > > > +}
> > > > +
> > > > +/*
> > > > + * Compute the maximum number of records (or keyptrs) per block that we want to
> > > > + * install at this level in the btree.  Caller is responsible for having set
> > > > + * @cur->bc_private.b.forksize to the desired fork size, if appropriate.
> > > > + */
> > > > +STATIC unsigned int
> > > > +xfs_btree_bload_max_npb(
> > > > +	struct xfs_btree_cur	*cur,
> > > > +	struct xfs_btree_bload	*bbl,
> > > > +	unsigned int		level)
> > > > +{
> > > > +	unsigned int		ret;
> > > > +
> > > > +	if (level == cur->bc_nlevels - 1 && cur->bc_ops->get_dmaxrecs)
> > > > +		return cur->bc_ops->get_dmaxrecs(cur, level);
> > > > +
> > > > +	ret = cur->bc_ops->get_maxrecs(cur, level);
> > > > +	if (level == 0)
> > > > +		ret -= bbl->leaf_slack;
> > > > +	else
> > > > +		ret -= bbl->node_slack;
> > > > +	return ret;
> > > > +}
> > > > +
> > > > +/*
> > > > + * Compute the desired number of records (or keyptrs) per block that we want to
> > > > + * install at this level in the btree, which must be somewhere between minrecs
> > > > + * and max_npb.  The caller is free to install fewer records per block.
> > > > + */
> > > > +STATIC unsigned int
> > > > +xfs_btree_bload_desired_npb(
> > > > +	struct xfs_btree_cur	*cur,
> > > > +	struct xfs_btree_bload	*bbl,
> > > > +	unsigned int		level)
> > > > +{
> > > > +	unsigned int		npb = xfs_btree_bload_max_npb(cur, bbl, level);
> > > > +
> > > > +	/* Root blocks are not subject to minrecs rules. */
> > > > +	if (level == cur->bc_nlevels - 1)
> > > > +		return max(1U, npb);
> > > > +
> > > > +	return max_t(unsigned int, cur->bc_ops->get_minrecs(cur, level), npb);
> > > > +}
> > > > +
> > > > +/*
> > > > + * Compute the number of records to be stored in each block at this level and
> > > > + * the number of blocks for this level.  For leaf levels, we must populate an
> > > > + * empty root block even if there are no records, so we have to have at least
> > > > + * one block.
> > > > + */
> > > > +STATIC void
> > > > +xfs_btree_bload_level_geometry(
> > > > +	struct xfs_btree_cur	*cur,
> > > > +	struct xfs_btree_bload	*bbl,
> > > > +	unsigned int		level,
> > > > +	uint64_t		nr_this_level,
> > > > +	unsigned int		*avg_per_block,
> > > > +	uint64_t		*blocks,
> > > > +	uint64_t		*blocks_with_extra)
> > > > +{
> > > > +	uint64_t		npb;
> > > > +	uint64_t		dontcare;
> > > > +	unsigned int		desired_npb;
> > > > +	unsigned int		maxnr;
> > > > +
> > > > +	maxnr = cur->bc_ops->get_maxrecs(cur, level);
> > > > +
> > > > +	/*
> > > > +	 * Compute the number of blocks we need to fill each block with the
> > > > +	 * desired number of records/keyptrs per block.  Because desired_npb
> > > > +	 * could be minrecs, we use regular integer division (which rounds
> > > > +	 * the block count down) so that in the next step the effective # of
> > > > +	 * items per block will never be less than desired_npb.
> > > > +	 */
> > > > +	desired_npb = xfs_btree_bload_desired_npb(cur, bbl, level);
> > > > +	*blocks = div64_u64_rem(nr_this_level, desired_npb, &dontcare);
> > > > +	*blocks = max(1ULL, *blocks);
> > > > +
> > > > +	/*
> > > > +	 * Compute the number of records that we will actually put in each
> > > > +	 * block, assuming that we want to spread the records evenly between
> > > > +	 * the blocks.  Take care that the effective # of items per block (npb)
> > > > +	 * won't exceed maxrecs even for the blocks that get an extra record,
> > > > +	 * since desired_npb could be maxrecs, and in the previous step we
> > > > +	 * rounded the block count down.
> > > > +	 */
> > > > +	npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > > +	if (npb > maxnr || (npb == maxnr && *blocks_with_extra > 0)) {
> > > > +		(*blocks)++;
> > > > +		npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > > +	}
> > > > +
> > > > +	*avg_per_block = min_t(uint64_t, npb, nr_this_level);
> > > > +
> > > > +	trace_xfs_btree_bload_level_geometry(cur, level, nr_this_level,
> > > > +			*avg_per_block, desired_npb, *blocks,
> > > > +			*blocks_with_extra);
> > > > +}
> > > > +
> > > > +/*
> > > > + * Ensure a slack value is appropriate for the btree.
> > > > + *
> > > > + * If the slack value is negative, set slack so that we fill the block to
> > > > + * halfway between minrecs and maxrecs.  Make sure the slack is never so large
> > > > + * that we can underflow minrecs.
> > > > + */
> > > > +static void
> > > > +xfs_btree_bload_ensure_slack(
> > > > +	struct xfs_btree_cur	*cur,
> > > > +	int			*slack,
> > > > +	int			level)
> > > > +{
> > > > +	int			maxr;
> > > > +	int			minr;
> > > > +
> > > > +	/*
> > > > +	 * We only care about slack for btree blocks, so set the btree nlevels
> > > > +	 * to 3 so that level 0 is a leaf block and level 1 is a node block.
> > > > +	 * Avoid straying into inode roots, since we don't do slack there.
> > > > +	 */
> > > > +	cur->bc_nlevels = 3;
> > > > +	maxr = cur->bc_ops->get_maxrecs(cur, level);
> > > > +	minr = cur->bc_ops->get_minrecs(cur, level);
> > > > +
> > > > +	/*
> > > > +	 * If slack is negative, automatically set slack so that we load the
> > > > +	 * btree block approximately halfway between minrecs and maxrecs.
> > > > +	 * Generally, this will net us 75% loading.
> > > > +	 */
> > > > +	if (*slack < 0)
> > > > +		*slack = maxr - ((maxr + minr) >> 1);
> > > > +
> > > > +	*slack = min(*slack, maxr - minr);
> > > > +}
> > > > +
> > > > +/*
> > > > + * Prepare a btree cursor for a bulk load operation by computing the geometry
> > > > + * fields in @bbl.  Caller must ensure that the btree cursor is a staging
> > > > + * cursor.  This function can be called multiple times.
> > > > + */
> > > > +int
> > > > +xfs_btree_bload_compute_geometry(
> > > > +	struct xfs_btree_cur	*cur,
> > > > +	struct xfs_btree_bload	*bbl,
> > > > +	uint64_t		nr_records)
> > > > +{
> > > > +	uint64_t		nr_blocks = 0;
> > > > +	uint64_t		nr_this_level;
> > > > +
> > > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > > +
> > > > +	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
> > > > +	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
> > > > +
> > > > +	bbl->nr_records = nr_this_level = nr_records;
> > > > +	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
> > > > +		uint64_t	level_blocks;
> > > > +		uint64_t	dontcare64;
> > > > +		unsigned int	level = cur->bc_nlevels - 1;
> > > > +		unsigned int	avg_per_block;
> > > > +
> > > > +		/*
> > > > +		 * If all the things we want to store at this level would fit
> > > > +		 * in a single root block, then we have our btree root and are
> > > > +		 * done.  Note that bmap btrees do not allow records in the
> > > > +		 * root.
> > > > +		 */
> > > > +		if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) || level != 0) {
> > > > +			xfs_btree_bload_level_geometry(cur, bbl, level,
> > > > +					nr_this_level, &avg_per_block,
> > > > +					&level_blocks, &dontcare64);
> > > > +			if (nr_this_level <= avg_per_block) {
> > > > +				nr_blocks++;
> > > > +				break;
> > > > +			}
> > > > +		}
> > > > +
> > > > +		/*
> > > > +		 * Otherwise, we have to store all the records for this level
> > > > +		 * in blocks and therefore need another level of btree to point
> > > > +		 * to those blocks.  Increase the number of levels and
> > > > +		 * recompute the number of records we can store at this level
> > > > +		 * because that can change depending on whether or not a level
> > > > +		 * is the root level.
> > > > +		 */
> > > > +		cur->bc_nlevels++;
> > > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > +				&avg_per_block, &level_blocks, &dontcare64);
> > > > +		nr_blocks += level_blocks;
> > > > +		nr_this_level = level_blocks;
> > > > +	}
> > > > +
> > > > +	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
> > > > +		return -EOVERFLOW;
> > > > +
> > > > +	bbl->btree_height = cur->bc_nlevels;
> > > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
> > > > +		bbl->nr_blocks = nr_blocks - 1;
> > > > +	else
> > > > +		bbl->nr_blocks = nr_blocks;
> > > > +	return 0;
> > > > +}
> > > > +
> > > > +/*
> > > > + * Bulk load a btree.
> > > > + *
> > > > + * Load @bbl->nr_records quantity of records into a btree using the supplied
> > > > + * empty and staging btree cursor @cur and a @bbl that has been filled out by
> > > > + * the xfs_btree_bload_compute_geometry function.
> > > > + *
> > > > + * The @bbl->get_data function must populate the cursor's bc_rec every time it
> > > > + * is called.  The @bbl->alloc_block function will be used to allocate new
> > > > + * btree blocks.  @priv is passed to both functions.
> > > > + *
> > > > + * Caller must ensure that @cur is a staging cursor.  Any existing btree rooted
> > > > + * in the fakeroot will be lost, so do not call this function twice.
> > > > + */
> > > > +int
> > > > +xfs_btree_bload(
> > > > +	struct xfs_btree_cur		*cur,
> > > > +	struct xfs_btree_bload		*bbl,
> > > > +	void				*priv)
> > > > +{
> > > > +	union xfs_btree_ptr		child_ptr;
> > > > +	union xfs_btree_ptr		ptr;
> > > > +	struct xfs_buf			*bp = NULL;
> > > > +	struct xfs_btree_block		*block = NULL;
> > > > +	uint64_t			nr_this_level = bbl->nr_records;
> > > > +	uint64_t			blocks;
> > > > +	uint64_t			i;
> > > > +	uint64_t			blocks_with_extra;
> > > > +	uint64_t			total_blocks = 0;
> > > > +	unsigned int			avg_per_block;
> > > > +	unsigned int			level = 0;
> > > > +	int				ret;
> > > > +
> > > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > > +
> > > > +	cur->bc_nlevels = bbl->btree_height;
> > > > +	xfs_btree_set_ptr_null(cur, &child_ptr);
> > > > +	xfs_btree_set_ptr_null(cur, &ptr);
> > > > +
> > > > +	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > +			&avg_per_block, &blocks, &blocks_with_extra);
> > > > +
> > > > +	/* Load each leaf block. */
> > > > +	for (i = 0; i < blocks; i++) {
> > > > +		unsigned int		nr_this_block = avg_per_block;
> > > > +
> > > > +		if (i < blocks_with_extra)
> > > > +			nr_this_block++;
> > > > +
> > > > +		ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > > +				nr_this_block, &ptr, &bp, &block, priv);
> > > > +		if (ret)
> > > > +			return ret;
> > > > +
> > > > +		trace_xfs_btree_bload_block(cur, level, i, blocks, &ptr,
> > > > +				nr_this_block);
> > > > +
> > > > +		ret = xfs_btree_bload_leaf(cur, nr_this_block, bbl->get_data,
> > > > +				block, priv);
> > > > +		if (ret)
> > > > +			goto out;
> > > > +
> > > > +		/* Record the leftmost pointer to start the next level. */
> > > > +		if (i == 0)
> > > > +			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
> > > > +	}
> > > > +	total_blocks += blocks;
> > > > +	xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > > +
> > > > +	/* Populate the internal btree nodes. */
> > > > +	for (level = 1; level < cur->bc_nlevels; level++) {
> > > > +		union xfs_btree_ptr	first_ptr;
> > > > +
> > > > +		nr_this_level = blocks;
> > > > +		block = NULL;
> > > > +		xfs_btree_set_ptr_null(cur, &ptr);
> > > > +
> > > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > +				&avg_per_block, &blocks, &blocks_with_extra);
> > > > +
> > > > +		/* Load each node block. */
> > > > +		for (i = 0; i < blocks; i++) {
> > > > +			unsigned int	nr_this_block = avg_per_block;
> > > > +
> > > > +			if (i < blocks_with_extra)
> > > > +				nr_this_block++;
> > > > +
> > > > +			ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > > +					nr_this_block, &ptr, &bp, &block,
> > > > +					priv);
> > > > +			if (ret)
> > > > +				return ret;
> > > > +
> > > > +			trace_xfs_btree_bload_block(cur, level, i, blocks,
> > > > +					&ptr, nr_this_block);
> > > > +
> > > > +			ret = xfs_btree_bload_node(cur, nr_this_block,
> > > > +					&child_ptr, block);
> > > > +			if (ret)
> > > > +				goto out;
> > > > +
> > > > +			/*
> > > > +			 * Record the leftmost pointer to start the next level.
> > > > +			 */
> > > > +			if (i == 0)
> > > > +				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
> > > > +		}
> > > > +		total_blocks += blocks;
> > > > +		xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > > +		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
> > > > +	}
> > > > +
> > > > +	/* Initialize the new root. */
> > > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
> > > > +		ASSERT(xfs_btree_ptr_is_null(cur, &ptr));
> > > > +		cur->bc_private.b.ifake->if_levels = cur->bc_nlevels;
> > > > +		cur->bc_private.b.ifake->if_blocks = total_blocks - 1;
> > > > +	} else {
> > > > +		cur->bc_private.a.afake->af_root = be32_to_cpu(ptr.s);
> > > > +		cur->bc_private.a.afake->af_levels = cur->bc_nlevels;
> > > > +		cur->bc_private.a.afake->af_blocks = total_blocks;
> > > > +	}
> > > > +out:
> > > > +	if (bp)
> > > > +		xfs_trans_brelse(cur->bc_tp, bp);
> > > > +	return ret;
> > > > +}
> > > > diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
> > > > index a17becb72ab8..5c6992a04ea2 100644
> > > > --- a/fs/xfs/libxfs/xfs_btree.h
> > > > +++ b/fs/xfs/libxfs/xfs_btree.h
> > > > @@ -582,4 +582,47 @@ void xfs_btree_stage_ifakeroot(struct xfs_btree_cur *cur,
> > > >  void xfs_btree_commit_ifakeroot(struct xfs_btree_cur *cur, int whichfork,
> > > >  		const struct xfs_btree_ops *ops);
> > > >  
> > > > +typedef int (*xfs_btree_bload_get_fn)(struct xfs_btree_cur *cur, void *priv);
> > > > +typedef int (*xfs_btree_bload_alloc_block_fn)(struct xfs_btree_cur *cur,
> > > > +		union xfs_btree_ptr *ptr, void *priv);
> > > > +typedef size_t (*xfs_btree_bload_iroot_size_fn)(struct xfs_btree_cur *cur,
> > > > +		unsigned int nr_this_level, void *priv);
> > > > +
> > > > +/* Bulk loading of staged btrees. */
> > > > +struct xfs_btree_bload {
> > > > +	/* Function to store a record in the cursor. */
> > > > +	xfs_btree_bload_get_fn		get_data;
> > > > +
> > > > +	/* Function to allocate a block for the btree. */
> > > > +	xfs_btree_bload_alloc_block_fn	alloc_block;
> > > > +
> > > > +	/* Function to compute the size of the in-core btree root block. */
> > > > +	xfs_btree_bload_iroot_size_fn	iroot_size;
> > > > +
> > > > +	/* Number of records the caller wants to store. */
> > > > +	uint64_t			nr_records;
> > > > +
> > > > +	/* Number of btree blocks needed to store those records. */
> > > > +	uint64_t			nr_blocks;
> > > > +
> > > > +	/*
> > > > +	 * Number of free records to leave in each leaf block.  If this (or
> > > > +	 * any of the slack values) are negative, this will be computed to
> > > > +	 * be halfway between maxrecs and minrecs.  This typically leaves the
> > > > +	 * block 75% full.
> > > > +	 */
> > > > +	int				leaf_slack;
> > > > +
> > > > +	/* Number of free keyptrs to leave in each node block. */
> > > > +	int				node_slack;
> > > > +
> > > > +	/* Computed btree height. */
> > > > +	unsigned int			btree_height;
> > > > +};
> > > > +
> > > > +int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
> > > > +		struct xfs_btree_bload *bbl, uint64_t nr_records);
> > > > +int xfs_btree_bload(struct xfs_btree_cur *cur, struct xfs_btree_bload *bbl,
> > > > +		void *priv);
> > > > +
> > > >  #endif	/* __XFS_BTREE_H__ */
> > > > diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
> > > > index bc85b89f88ca..9b5e58a92381 100644
> > > > --- a/fs/xfs/xfs_trace.c
> > > > +++ b/fs/xfs/xfs_trace.c
> > > > @@ -6,6 +6,7 @@
> > > >  #include "xfs.h"
> > > >  #include "xfs_fs.h"
> > > >  #include "xfs_shared.h"
> > > > +#include "xfs_bit.h"
> > > >  #include "xfs_format.h"
> > > >  #include "xfs_log_format.h"
> > > >  #include "xfs_trans_resv.h"
> > > > diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
> > > > index a78055521fcd..6d7ba64b7a0f 100644
> > > > --- a/fs/xfs/xfs_trace.h
> > > > +++ b/fs/xfs/xfs_trace.h
> > > > @@ -35,6 +35,7 @@ struct xfs_icreate_log;
> > > >  struct xfs_owner_info;
> > > >  struct xfs_trans_res;
> > > >  struct xfs_inobt_rec_incore;
> > > > +union xfs_btree_ptr;
> > > >  
> > > >  DECLARE_EVENT_CLASS(xfs_attr_list_class,
> > > >  	TP_PROTO(struct xfs_attr_list_context *ctx),
> > > > @@ -3670,6 +3671,90 @@ TRACE_EVENT(xfs_btree_commit_ifakeroot,
> > > >  		  __entry->blocks)
> > > >  )
> > > >  
> > > > +TRACE_EVENT(xfs_btree_bload_level_geometry,
> > > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > > +		 uint64_t nr_this_level, unsigned int nr_per_block,
> > > > +		 unsigned int desired_npb, uint64_t blocks,
> > > > +		 uint64_t blocks_with_extra),
> > > > +	TP_ARGS(cur, level, nr_this_level, nr_per_block, desired_npb, blocks,
> > > > +		blocks_with_extra),
> > > > +	TP_STRUCT__entry(
> > > > +		__field(dev_t, dev)
> > > > +		__field(xfs_btnum_t, btnum)
> > > > +		__field(unsigned int, level)
> > > > +		__field(unsigned int, nlevels)
> > > > +		__field(uint64_t, nr_this_level)
> > > > +		__field(unsigned int, nr_per_block)
> > > > +		__field(unsigned int, desired_npb)
> > > > +		__field(unsigned long long, blocks)
> > > > +		__field(unsigned long long, blocks_with_extra)
> > > > +	),
> > > > +	TP_fast_assign(
> > > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > > +		__entry->btnum = cur->bc_btnum;
> > > > +		__entry->level = level;
> > > > +		__entry->nlevels = cur->bc_nlevels;
> > > > +		__entry->nr_this_level = nr_this_level;
> > > > +		__entry->nr_per_block = nr_per_block;
> > > > +		__entry->desired_npb = desired_npb;
> > > > +		__entry->blocks = blocks;
> > > > +		__entry->blocks_with_extra = blocks_with_extra;
> > > > +	),
> > > > +	TP_printk("dev %d:%d btree %s level %u/%u nr_this_level %llu nr_per_block %u desired_npb %u blocks %llu blocks_with_extra %llu",
> > > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > > +		  __entry->level,
> > > > +		  __entry->nlevels,
> > > > +		  __entry->nr_this_level,
> > > > +		  __entry->nr_per_block,
> > > > +		  __entry->desired_npb,
> > > > +		  __entry->blocks,
> > > > +		  __entry->blocks_with_extra)
> > > > +)
> > > > +
> > > > +TRACE_EVENT(xfs_btree_bload_block,
> > > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > > +		 uint64_t block_idx, uint64_t nr_blocks,
> > > > +		 union xfs_btree_ptr *ptr, unsigned int nr_records),
> > > > +	TP_ARGS(cur, level, block_idx, nr_blocks, ptr, nr_records),
> > > > +	TP_STRUCT__entry(
> > > > +		__field(dev_t, dev)
> > > > +		__field(xfs_btnum_t, btnum)
> > > > +		__field(unsigned int, level)
> > > > +		__field(unsigned long long, block_idx)
> > > > +		__field(unsigned long long, nr_blocks)
> > > > +		__field(xfs_agnumber_t, agno)
> > > > +		__field(xfs_agblock_t, agbno)
> > > > +		__field(unsigned int, nr_records)
> > > > +	),
> > > > +	TP_fast_assign(
> > > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > > +		__entry->btnum = cur->bc_btnum;
> > > > +		__entry->level = level;
> > > > +		__entry->block_idx = block_idx;
> > > > +		__entry->nr_blocks = nr_blocks;
> > > > +		if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
> > > > +			xfs_fsblock_t	fsb = be64_to_cpu(ptr->l);
> > > > +
> > > > +			__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsb);
> > > > +			__entry->agbno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsb);
> > > > +		} else {
> > > > +			__entry->agno = cur->bc_private.a.agno;
> > > > +			__entry->agbno = be32_to_cpu(ptr->s);
> > > > +		}
> > > > +		__entry->nr_records = nr_records;
> > > > +	),
> > > > +	TP_printk("dev %d:%d btree %s level %u block %llu/%llu fsb (%u/%u) recs %u",
> > > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > > +		  __entry->level,
> > > > +		  __entry->block_idx,
> > > > +		  __entry->nr_blocks,
> > > > +		  __entry->agno,
> > > > +		  __entry->agbno,
> > > > +		  __entry->nr_records)
> > > > +)
> > > > +
> > > >  #endif /* _TRACE_XFS_H */
> > > >  
> > > >  #undef TRACE_INCLUDE_PATH
> > > >

Brian Foster Oct. 17, 2019, 9:32 a.m. UTC | #5

On Wed, Oct 16, 2019 at 05:40:18PM -0700, Darrick J. Wong wrote:
> On Wed, Oct 16, 2019 at 05:07:31PM -0400, Brian Foster wrote:
> > On Wed, Oct 16, 2019 at 11:15:02AM -0700, Darrick J. Wong wrote:
> > > On Wed, Oct 16, 2019 at 11:26:48AM -0400, Brian Foster wrote:
> > > > On Wed, Oct 09, 2019 at 09:48:18AM -0700, Darrick J. Wong wrote:
> > > > > From: Darrick J. Wong <[e-mail]>
> > > > > 
> > > > > Add a new btree function that enables us to bulk load a btree cursor.
> > > > > This will be used by the upcoming online repair patches to generate new
> > > > > btrees.
> > > > > 
> > > > > Signed-off-by: Darrick J. Wong <[e-mail]>
> > > > > ---
> > > > >  fs/xfs/libxfs/xfs_btree.c |  566 +++++++++++++++++++++++++++++++++++++++++++++
> > > > >  fs/xfs/libxfs/xfs_btree.h |   43 +++
> > > > >  fs/xfs/xfs_trace.c        |    1 
> > > > >  fs/xfs/xfs_trace.h        |   85 +++++++
> > > > >  4 files changed, 694 insertions(+), 1 deletion(-)
> > > > > 
> > > > > 
> > > > > diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
> > > > > index 4b06d5d86834..17b0fdb87729 100644
> > > > > --- a/fs/xfs/libxfs/xfs_btree.c
> > > > > +++ b/fs/xfs/libxfs/xfs_btree.c
> > > > ...
> > > > > @@ -5104,3 +5104,567 @@ xfs_btree_commit_ifakeroot(
> > > > >  	cur->bc_ops = ops;
> > > > >  	cur->bc_flags &= ~XFS_BTREE_STAGING;
> > > > >  }
> > > > > +
> > > > > +/*
> > > > > + * Bulk Loading of Staged Btrees
> > > > > + * =============================
> > > > > + *
> > > > > + * This interface is used with a staged btree cursor to create a totally new
> > > > > + * btree with a large number of records (i.e. more than what would fit in a
> > > > > + * single block).  When the creation is complete, the new root can be linked
> > > > > + * atomically into the filesystem by committing the staged cursor.
> > > > > + *
> > > 
> > > [paraphrasing a conversation we had on irc]
> > > 
> > > > Thanks for the documentation. So what is the purpose behind the whole
> > > > bulk loading thing as opposed to something like faking up an AG
> > > > structure (i.e. AGF) somewhere and using the existing cursor mechanisms
> > > > (or something closer to it) to copy records from one place to another?
> > > > Is it purely a performance/efficiency tradeoff? Bulk block allocation
> > > > issues? Transactional/atomicity issues? All (or none :P) of the above?
> > > 
> > > Prior to the v20, the online repair series created a new btree root,
> > > committed that into wherever the root lived, and inserted records one by
> > > one into the btree.  There were quite a few drawbacks to this method:
> > > 
> > > 1. Inserting records one at a time can involve walking up the tree to
> > > update node block pointers, which isn't terribly efficient if we're
> > > likely going to rewrite the pointers (and relogging nodes) several more
> > > times.
> > > 
> > > 2. Inserting records one at a time tends to leave a lot of half-empty
> > > btree blocks because when one block fills up we split it and push half
> > > the records to the new block.  It would be nice not to explode the size
> > > of the btrees, and it would be particularly useful if we could control
> > > the load factor of the new btree precisely.
> > > 
> > 
> > Interesting... this is a trait the traditional btree update paths share
> > though, right?
> 
> Right.  It's similar to the behavior Dave was seeing a couple of weeks
> ago with Zorro's stress testing of the incore extent cache.
> 
> > > 3. The rebuild wasn't atomic, since we were replacing the root prior to
> > > the insert loop.  If we crashed midway through a rebuild we'd end up
> > > with a garbage btree and no indication that it was incorrect.  That's
> > > how the fakeroot code got started.
> > > 
> > 
> > Indeed, though this seems more related to the anchoring (i.e. fake root)
> > approach than bulk vs. iterative construction.
> 
> Correct.
> 
> > > 4. In a previous version of the repair series I tried to batch as many
> > > insert operations into a single transaction as possible, but my
> > > transaction reservation fullness estimation function didn't work
> > > reliably (particularly when things got really fragmented), so I backed
> > > off to rolling after /every/ insertion.  That works well enough, but at
> > > a cost of a lot of transaction rolling, which means that repairs plod
> > > along very slowly.
> > > 
> > > 5. Performing an insert loop means that the btree blocks are allocated
> > > one at a time as the btree expands.  This is suboptimal since we can
> > > calculate the exact size of the new btree prior to building it, which
> > > gives us the opportunity to recreate the index in a set of contiguous
> > > blocks instead of scattering them.
> > > 
> > 
> > Yep, FWIW it sounds like most of these tradeoffs are around
> > performance/efficiency. 
> 
> <nod>
> 
> > > 6. If we crash midway through a rebuild, XFS neither cleaned up the mess
> > > nor informed the administrator that it was necessary to re-run xfs_scrub
> > > or xfs_repair to clean up the lost blocks.  Obviously, automatic cleanup
> > > is a far better solution.
> > > 
> > 
> > Similar to above, I think this kind of depends more on how/where to
> > anchor an in-progress tree as opposed to what level records are copied
> > at.
> 
> <nod> The six points are indeed the overall list of complaints about the
> v19 code. :)
> 
> > > The first thing I decided to solve was the lack of atomicity.
> > > 
> > > For AG-rooted btrees, I thought about creating a fake xfs_buf for an AG
> > > header buffer and extracting the root/level values after construction
> > > completes.  That's possible, but it's risky because the fake buffer
> > > could get logged and if the sector number matches the actual header
> > > then it introduces buffer cache aliasing issues.
> > > 
> > > For inode-rooted btrees, one could create a fake xfs_inode with the same
> > > i_ino as the target.  That presents the same aliasing issues as the fake
> > > xfs_buf above.  A different strategy would be to allocate an unlinked
> > > inode and then use the bmbt owner change (a.k.a. extent swap) to move
> > > the mappings over.  That would work, though it has two large drawbacks:
> > > (a) a lot of additional complexity around allocating and freeing the
> > > temporary inode; and (b) future inode-rooted btrees such as the realtime
> > > rmap btree would also have to implement an owner-change operation.
> > > 
> > 
> > I was wondering more along the lines of having an actual anchor
> > somewhere. E.g., think of it as a temporary/inaccessible location of a
> > legitimate on-disk structure as opposed to a fake object in memory
> > somewhere. A hidden/internal repair inode or some such, perhaps. I'm
> > sure there's new code/complexity that would come around with that, but I
> > think that's going to be unavoidable to some degree for an online repair
> > mechanism. ;)
> 
> <nod> So far I /think/ I've managed to keep to an absolute minimum the
> amount of metadata that gets written to disk prior to the commit.  I
> haven't reread the series with an eye for how v20 is going to come up
> short though. :)
> 

This metadata has to be written out one way or another though. IOW, it
seems the advantage to holding off is mostly for the case where we crash
or something mid reconstruction (which should be the uncommon case :P).
I could see how there might be some benefits to using contiguous blocks
in a btree and batching them out as such, but conversely I'm also
curious about how that might behave with certain scenarios like really
large btrees that take a while to reconstruct (e.g. due to read
contention or slow storage) or where contiguous blocks aren't available,
etc.

> I may very well have to revisit the hidden/internal repair inode concept
> whenever I start working on rebuilding directories and xattrs since I
> can't see any other way of atomically rebuilding those.  But that's
> very very far out still.
> 

Ok.

> > Note that this is all just handwaving on my part and still without full
> > context as to how things are currently anchored, made atomic, etc. I'm
> > primarily trying to understand the design reasoning based on the high
> > level description.
> 
> <nod>
> 
> > > To fix (3), I thought it wise to have explicit fakeroot structures to
> > > maintain a clean separation between what we're building and the rest of
> > > the filesystem.  This also means that there's nothing on disk to clean
> > > up if we fail at any point before we're ready to commit the new btree.
> > > 
> > 
> > Hmm.. so this approach facilites a tree reconstruction in a single open
> > transaction? If so, I suppose I could see some functional advantages to
> > that.
> 
> Correct.
> 
> > > Then Dave (I think?) suggested that I  use EFIs strategically to
> > > schedule freeing of the new btree blocks (the root commit transaction
> > > would log EFDs to cancel them) and to schedule freeing of the old
> > > blocks.  That solves (6), though the EFI wrangling doesn't happen for
> > > another couple of series after this one.
> > > 
> > 
> > Hm, Ok... so new btree block allocation(s?) in the same transaction as
> > an EFI, to be processed on recovery if we crash, otherwise cancelled
> > with an EFD on construction completion..?
> 
> Correct.  In the end, the transaction sequence looks like:
> 
> T[1]: Allocate an extent, log metadata updates to reflect that, log EFI
> for the extent.
> 
> <repeat until we've allocated as many blocks as we need>
> 

If I follow correctly, I take it "repeat" here means we're rolling the
transaction to reallocate, and relog all previous EFIs along the way,
until we've acquired sufficient blocks to start populating the tree.

> T[N]: Attach ordered buffers for the new btree's blocks.  Log the root
> change.  Log EFDs for all the EFIs logged in T[1..N-1].  Log EFIs for
> all the old btree blocks that we could find.
> 

So here we're on/after the last allocation transaction, have a bunch of
pending EFIs for recovery purposes, and start populating the btree
blocks...

AFAICT the log tail is pinned by these EFIs during population, which is
of non-deterministic time for aforementioned reasons. If so, this
_sounds_ like it risks a similar log deadlock vector to the one we
currently have for quotaoff, though I could easily be missing something
from the high level description.

Also, what happens if we complete the population, roll this transaction,
unpin the ordered buffers and crash before the entire set of ordered
buffers is written back? We currently use ordered buffers with
operations described by logical intents. For example, an icreate item
goes into the log with an ordered buffer such that if we crash before
buffer writeback completes and allows the tail to move, recovery knows
how to turn the active icreate log item back into a valid inode chunk.
Is there something similar going on here?

> <roll transaction to write the ordered buffers and commit>
> 
> T[N+1]: Free an extent to finish the first EFI logged in the previous step.
> 
> <repeat until we've processed everything from the second wave of EFIs>
> 

And I take repeat here means essentially rolling and completing the
"real" EFIs for the old btree we've swapped out.

FWIW, it seems to me that this altogether might be an opportunity for a
custom logical intent to cover reconstruction sequences. 

> Call xfs_trans_commit and we're done.
> 
> > > He also suggested using ordered buffers to write out the new btree
> > > blocks along with whatever logging was necessary to commit the new
> > > btree.  It then occurred to me that xfs_repair open-codes the process of
> > > calculating the geometry of a new btree, allocating all the blocks at
> > > once, and writing out full btree blocks.  Somewhat annoyingly, it
> > > features nearly the same (open-)code for all four AG btree types, which
> > > is less maintainable than it could be.
> > > 
> > > I read through all four versions and used it to write the generic btree
> > > bulk loading code.  For scrub I hooked that up to the "staged btree with
> > > a fake root" stuff I'd already written, which solves (1), (2), (4), and
> > > (5).
> > > 
> > > For xfsprogs[1], I deleted a few thousand lines of code from xfs_repair.
> > > True, we don't reuse existing common code, but we at least get to share
> > > new common btree code.
> > > 
> > 
> > Yeah, the xfsprogs work certainly makes sense. Part of the reason I ask
> > about this is the tradeoff of having multiple avenues to construct a
> > tree in the kernel codebase.
> 
> <nod>
> 
> > > > This is my first pass through this so I'm mostly looking at big picture
> > > > until I get to a point to see how these bits are used. The mechanism
> > > > itself seems reasonable in principle, but the reason I ask is it also
> > > > seems like there's inherent value in using more of same infrastructure
> > > > to reconstruct a tree that we use to create one in the first place. We
> > > > also already have primitives for things like fork swapping via the
> > > > extent swap mechanism, etc.
> > > 
> > > "bfoster: I guess it would be nice to see that kind of make it work ->
> > > make it fast evolution in tree"
> > > 
> > > For a while I did maintain the introduction of the bulk loading code as
> > > separate patches against the v19 repair code, but unfortunately I
> > > smushed them down before sending v20 to reduce the patch count, and
> > > because I didn't want to argue with everyone over the semi-working code
> > > that would then be replaced in the very next patch.
> > > 
> > 
> > That's not quite what I meant... The approach you've taken makes sense
> > to me for an implementation presented in a single series. I was more
> > thinking that at the point where it was determined the implementation
> > was going to change so drastically, after so many iterations it might
> > have been useful to see the v19 approach merged in an experimental form
> 
> I would have liked to see the online repair stuff merged in experimental
> form too so I can reduce the size of my patch queue, but oh well. :)
> 

Oh well, indeed. :P

> The silver lining to these lengthy reworks and slow review is that I can
> come back and do a fresh self-review after a month and straighten out
> the ugly parts as time goes by.  Unfortunately, that doesn't leave much
> of a paper trail or obvious evidence of development history.
> 

Exactly.

> Eight months ago it occurred to me that perhaps there is some value in
> retaining *some* periodic development history of this, so I've been
> adding dated tags to my integration repo[1] based on my development
> branch names, so I guess you could actually clone the git repo and git
> diff from one tag to another.  In general I'll generate a new pile of
> tags just before patchbombing.
> 

More importantly, I think we should try to follow that approach if
possible from here on, even if there are open/unresolved issues with the
currently proposed approach. Online repair overall is complex enough
that it would probably be impossible to get a fully functional online
repair done in a single series (and we're clearly not taking that
approach anyways), making sure that the various common underpinnings
work for all of the various types of structures, etc. Just my .02
though..

Brian

> (Granted 'repair-part-one' has been split into smaller parts now...)
> 
> > and then reworked upstream from there. Now that the new approach is
> > implemented, I agree it's probably not worth reinserting the old
> > approach at this point just to switch it out.
> 
> <nod>
> 
> > Thanks for the breakdown...
> 
> No problem, thanks for reading! :)
> 
> --D
> 
> [1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux.git
> 
> > Brian
> > 
> > > I could split them back out, though at a cost of having to reintroduce a
> > > lot of hairy code in the bnobt/cntbt rebuild function to seed the free
> > > new space btree root in order to make sure that the btree block
> > > allocation code works properly, along with auditing the allocation paths
> > > to make sure they don't use the old AGF or encounter other subtleties.
> > > 
> > > It'd be a lot of work considering that the v20 reconstruction code is
> > > /much/ simpler than v19's was.  I also restructured the repair functions
> > > to allocate one large context structure at the beginning instead of the
> > > piecemeal way it was done onstack in v19 because stack usage was growing
> > > close to 1k in some cases.
> > > 
> > > --D
> > > 
> > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/log/?h=repair-bulk-load
> > > 
> > > > 
> > > > Brian
> > > > 
> > > > > + * The first step for the caller is to construct a fake btree root structure
> > > > > + * and a staged btree cursor.  A staging cursor contains all the geometry
> > > > > + * information for the btree type but will fail all operations that could have
> > > > > + * side effects in the filesystem (e.g. btree shape changes).  Regular
> > > > > + * operations will not work unless the staging cursor is committed and becomes
> > > > > + * a regular cursor.
> > > > > + *
> > > > > + * For a btree rooted in an AG header, use an xbtree_afakeroot structure.
> > > > > + * This should be initialized to zero.  For a btree rooted in an inode fork,
> > > > > + * use an xbtree_ifakeroot structure.  @if_fork_size field should be set to
> > > > > + * the number of bytes available to the fork in the inode; @if_fork should
> > > > > + * point to a freshly allocated xfs_inode_fork; and @if_format should be set
> > > > > + * to the appropriate fork type (e.g. XFS_DINODE_FMT_BTREE).
> > > > > + *
> > > > > + * The next step for the caller is to initialize a struct xfs_btree_bload
> > > > > + * context.  The @nr_records field is the number of records that are to be
> > > > > + * loaded into the btree.  The @leaf_slack and @node_slack fields are the
> > > > > + * number of records (or key/ptr) slots to leave empty in new btree blocks.
> > > > > + * If a caller sets a slack value to -1, the slack value will be computed to
> > > > > + * fill the block halfway between minrecs and maxrecs items per block.
> > > > > + *
> > > > > + * The number of items placed in each btree block is computed via the following
> > > > > + * algorithm: For leaf levels, the number of items for the level is nr_records.
> > > > > + * For node levels, the number of items for the level is the number of blocks
> > > > > + * in the next lower level of the tree.  For each level, the desired number of
> > > > > + * items per block is defined as:
> > > > > + *
> > > > > + * desired = max(minrecs, maxrecs - slack factor)
> > > > > + *
> > > > > + * The number of blocks for the level is defined to be:
> > > > > + *
> > > > > + * blocks = nr_items / desired
> > > > > + *
> > > > > + * Note this is rounded down so that the npb calculation below will never fall
> > > > > + * below minrecs.  The number of items that will actually be loaded into each
> > > > > + * btree block is defined as:
> > > > > + *
> > > > > + * npb =  nr_items / blocks
> > > > > + *
> > > > > + * Some of the leftmost blocks in the level will contain one extra record as
> > > > > + * needed to handle uneven division.  If the number of records in any block
> > > > > + * would exceed maxrecs for that level, blocks is incremented and npb is
> > > > > + * recalculated.
> > > > > + *
> > > > > + * In other words, we compute the number of blocks needed to satisfy a given
> > > > > + * loading level, then spread the items as evenly as possible.
> > > > > + *
> > > > > + * To complete this step, call xfs_btree_bload_compute_geometry, which uses
> > > > > + * those settings to compute the height of the btree and the number of blocks
> > > > > + * that will be needed to construct the btree.  These values are stored in the
> > > > > + * @btree_height and @nr_blocks fields.
> > > > > + *
> > > > > + * At this point, the caller must allocate @nr_blocks blocks and save them for
> > > > > + * later.  If space is to be allocated transactionally, the staging cursor
> > > > > + * must be deleted before and recreated after, which is why computing the
> > > > > + * geometry is a separate step.
> > > > > + *
> > > > > + * The fourth step in the bulk loading process is to set the function pointers
> > > > > + * in the bload context structure.  @get_data will be called for each record
> > > > > + * that will be loaded into the btree; it should set the cursor's bc_rec
> > > > > + * field, which will be converted to on-disk format and copied into the
> > > > > + * appropriate record slot.  @alloc_block should supply one of the blocks
> > > > > + * allocated in the previous step.  For btrees which are rooted in an inode
> > > > > + * fork, @iroot_size is called to compute the size of the incore btree root
> > > > > + * block.  Call xfs_btree_bload to start constructing the btree.
> > > > > + *
> > > > > + * The final step is to commit the staging cursor, which logs the new btree
> > > > > + * root and turns the btree into a regular btree cursor, and free the fake
> > > > > + * roots.
> > > > > + */
> > > > > +
> > > > > +/*
> > > > > + * Put a btree block that we're loading onto the ordered list and release it.
> > > > > + * The btree blocks will be written when the final transaction swapping the
> > > > > + * btree roots is committed.
> > > > > + */
> > > > > +static void
> > > > > +xfs_btree_bload_drop_buf(
> > > > > +	struct xfs_trans	*tp,
> > > > > +	struct xfs_buf		**bpp)
> > > > > +{
> > > > > +	if (*bpp == NULL)
> > > > > +		return;
> > > > > +
> > > > > +	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_BTREE_BUF);
> > > > > +	xfs_trans_ordered_buf(tp, *bpp);
> > > > > +	xfs_trans_brelse(tp, *bpp);
> > > > > +	*bpp = NULL;
> > > > > +}
> > > > > +
> > > > > +/* Allocate and initialize one btree block for bulk loading. */
> > > > > +STATIC int
> > > > > +xfs_btree_bload_prep_block(
> > > > > +	struct xfs_btree_cur		*cur,
> > > > > +	struct xfs_btree_bload		*bbl,
> > > > > +	unsigned int			level,
> > > > > +	unsigned int			nr_this_block,
> > > > > +	union xfs_btree_ptr		*ptrp,
> > > > > +	struct xfs_buf			**bpp,
> > > > > +	struct xfs_btree_block		**blockp,
> > > > > +	void				*priv)
> > > > > +{
> > > > > +	union xfs_btree_ptr		new_ptr;
> > > > > +	struct xfs_buf			*new_bp;
> > > > > +	struct xfs_btree_block		*new_block;
> > > > > +	int				ret;
> > > > > +
> > > > > +	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
> > > > > +	    level == cur->bc_nlevels - 1) {
> > > > > +		struct xfs_ifork	*ifp = cur->bc_private.b.ifake->if_fork;
> > > > > +		size_t			new_size;
> > > > > +
> > > > > +		/* Allocate a new incore btree root block. */
> > > > > +		new_size = bbl->iroot_size(cur, nr_this_block, priv);
> > > > > +		ifp->if_broot = kmem_zalloc(new_size, 0);
> > > > > +		ifp->if_broot_bytes = (int)new_size;
> > > > > +		ifp->if_flags |= XFS_IFBROOT;
> > > > > +
> > > > > +		/* Initialize it and send it out. */
> > > > > +		xfs_btree_init_block_int(cur->bc_mp, ifp->if_broot,
> > > > > +				XFS_BUF_DADDR_NULL, cur->bc_btnum, level,
> > > > > +				nr_this_block, cur->bc_private.b.ip->i_ino,
> > > > > +				cur->bc_flags);
> > > > > +
> > > > > +		*bpp = NULL;
> > > > > +		*blockp = ifp->if_broot;
> > > > > +		xfs_btree_set_ptr_null(cur, ptrp);
> > > > > +		return 0;
> > > > > +	}
> > > > > +
> > > > > +	/* Allocate a new leaf block. */
> > > > > +	ret = bbl->alloc_block(cur, &new_ptr, priv);
> > > > > +	if (ret)
> > > > > +		return ret;
> > > > > +
> > > > > +	ASSERT(!xfs_btree_ptr_is_null(cur, &new_ptr));
> > > > > +
> > > > > +	ret = xfs_btree_get_buf_block(cur, &new_ptr, &new_block, &new_bp);
> > > > > +	if (ret)
> > > > > +		return ret;
> > > > > +
> > > > > +	/* Initialize the btree block. */
> > > > > +	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
> > > > > +	if (*blockp)
> > > > > +		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
> > > > > +	xfs_btree_set_sibling(cur, new_block, ptrp, XFS_BB_LEFTSIB);
> > > > > +	xfs_btree_set_numrecs(new_block, nr_this_block);
> > > > > +
> > > > > +	/* Release the old block and set the out parameters. */
> > > > > +	xfs_btree_bload_drop_buf(cur->bc_tp, bpp);
> > > > > +	*blockp = new_block;
> > > > > +	*bpp = new_bp;
> > > > > +	xfs_btree_copy_ptrs(cur, ptrp, &new_ptr, 1);
> > > > > +	return 0;
> > > > > +}
> > > > > +
> > > > > +/* Load one leaf block. */
> > > > > +STATIC int
> > > > > +xfs_btree_bload_leaf(
> > > > > +	struct xfs_btree_cur		*cur,
> > > > > +	unsigned int			recs_this_block,
> > > > > +	xfs_btree_bload_get_fn		get_data,
> > > > > +	struct xfs_btree_block		*block,
> > > > > +	void				*priv)
> > > > > +{
> > > > > +	unsigned int			j;
> > > > > +	int				ret;
> > > > > +
> > > > > +	/* Fill the leaf block with records. */
> > > > > +	for (j = 1; j <= recs_this_block; j++) {
> > > > > +		union xfs_btree_rec	*block_recs;
> > > > > +
> > > > > +		ret = get_data(cur, priv);
> > > > > +		if (ret)
> > > > > +			return ret;
> > > > > +		block_recs = xfs_btree_rec_addr(cur, j, block);
> > > > > +		cur->bc_ops->init_rec_from_cur(cur, block_recs);
> > > > > +	}
> > > > > +
> > > > > +	return 0;
> > > > > +}
> > > > > +
> > > > > +/* Load one node block. */
> > > > > +STATIC int
> > > > > +xfs_btree_bload_node(
> > > > > +	struct xfs_btree_cur	*cur,
> > > > > +	unsigned int		recs_this_block,
> > > > > +	union xfs_btree_ptr	*child_ptr,
> > > > > +	struct xfs_btree_block	*block)
> > > > > +{
> > > > > +	unsigned int		j;
> > > > > +	int			ret;
> > > > > +
> > > > > +	/* Fill the node block with keys and pointers. */
> > > > > +	for (j = 1; j <= recs_this_block; j++) {
> > > > > +		union xfs_btree_key	child_key;
> > > > > +		union xfs_btree_ptr	*block_ptr;
> > > > > +		union xfs_btree_key	*block_key;
> > > > > +		struct xfs_btree_block	*child_block;
> > > > > +		struct xfs_buf		*child_bp;
> > > > > +
> > > > > +		ASSERT(!xfs_btree_ptr_is_null(cur, child_ptr));
> > > > > +
> > > > > +		ret = xfs_btree_get_buf_block(cur, child_ptr, &child_block,
> > > > > +				&child_bp);
> > > > > +		if (ret)
> > > > > +			return ret;
> > > > > +
> > > > > +		xfs_btree_get_keys(cur, child_block, &child_key);
> > > > > +
> > > > > +		block_ptr = xfs_btree_ptr_addr(cur, j, block);
> > > > > +		xfs_btree_copy_ptrs(cur, block_ptr, child_ptr, 1);
> > > > > +
> > > > > +		block_key = xfs_btree_key_addr(cur, j, block);
> > > > > +		xfs_btree_copy_keys(cur, block_key, &child_key, 1);
> > > > > +
> > > > > +		xfs_btree_get_sibling(cur, child_block, child_ptr,
> > > > > +				XFS_BB_RIGHTSIB);
> > > > > +		xfs_trans_brelse(cur->bc_tp, child_bp);
> > > > > +	}
> > > > > +
> > > > > +	return 0;
> > > > > +}
> > > > > +
> > > > > +/*
> > > > > + * Compute the maximum number of records (or keyptrs) per block that we want to
> > > > > + * install at this level in the btree.  Caller is responsible for having set
> > > > > + * @cur->bc_private.b.forksize to the desired fork size, if appropriate.
> > > > > + */
> > > > > +STATIC unsigned int
> > > > > +xfs_btree_bload_max_npb(
> > > > > +	struct xfs_btree_cur	*cur,
> > > > > +	struct xfs_btree_bload	*bbl,
> > > > > +	unsigned int		level)
> > > > > +{
> > > > > +	unsigned int		ret;
> > > > > +
> > > > > +	if (level == cur->bc_nlevels - 1 && cur->bc_ops->get_dmaxrecs)
> > > > > +		return cur->bc_ops->get_dmaxrecs(cur, level);
> > > > > +
> > > > > +	ret = cur->bc_ops->get_maxrecs(cur, level);
> > > > > +	if (level == 0)
> > > > > +		ret -= bbl->leaf_slack;
> > > > > +	else
> > > > > +		ret -= bbl->node_slack;
> > > > > +	return ret;
> > > > > +}
> > > > > +
> > > > > +/*
> > > > > + * Compute the desired number of records (or keyptrs) per block that we want to
> > > > > + * install at this level in the btree, which must be somewhere between minrecs
> > > > > + * and max_npb.  The caller is free to install fewer records per block.
> > > > > + */
> > > > > +STATIC unsigned int
> > > > > +xfs_btree_bload_desired_npb(
> > > > > +	struct xfs_btree_cur	*cur,
> > > > > +	struct xfs_btree_bload	*bbl,
> > > > > +	unsigned int		level)
> > > > > +{
> > > > > +	unsigned int		npb = xfs_btree_bload_max_npb(cur, bbl, level);
> > > > > +
> > > > > +	/* Root blocks are not subject to minrecs rules. */
> > > > > +	if (level == cur->bc_nlevels - 1)
> > > > > +		return max(1U, npb);
> > > > > +
> > > > > +	return max_t(unsigned int, cur->bc_ops->get_minrecs(cur, level), npb);
> > > > > +}
> > > > > +
> > > > > +/*
> > > > > + * Compute the number of records to be stored in each block at this level and
> > > > > + * the number of blocks for this level.  For leaf levels, we must populate an
> > > > > + * empty root block even if there are no records, so we have to have at least
> > > > > + * one block.
> > > > > + */
> > > > > +STATIC void
> > > > > +xfs_btree_bload_level_geometry(
> > > > > +	struct xfs_btree_cur	*cur,
> > > > > +	struct xfs_btree_bload	*bbl,
> > > > > +	unsigned int		level,
> > > > > +	uint64_t		nr_this_level,
> > > > > +	unsigned int		*avg_per_block,
> > > > > +	uint64_t		*blocks,
> > > > > +	uint64_t		*blocks_with_extra)
> > > > > +{
> > > > > +	uint64_t		npb;
> > > > > +	uint64_t		dontcare;
> > > > > +	unsigned int		desired_npb;
> > > > > +	unsigned int		maxnr;
> > > > > +
> > > > > +	maxnr = cur->bc_ops->get_maxrecs(cur, level);
> > > > > +
> > > > > +	/*
> > > > > +	 * Compute the number of blocks we need to fill each block with the
> > > > > +	 * desired number of records/keyptrs per block.  Because desired_npb
> > > > > +	 * could be minrecs, we use regular integer division (which rounds
> > > > > +	 * the block count down) so that in the next step the effective # of
> > > > > +	 * items per block will never be less than desired_npb.
> > > > > +	 */
> > > > > +	desired_npb = xfs_btree_bload_desired_npb(cur, bbl, level);
> > > > > +	*blocks = div64_u64_rem(nr_this_level, desired_npb, &dontcare);
> > > > > +	*blocks = max(1ULL, *blocks);
> > > > > +
> > > > > +	/*
> > > > > +	 * Compute the number of records that we will actually put in each
> > > > > +	 * block, assuming that we want to spread the records evenly between
> > > > > +	 * the blocks.  Take care that the effective # of items per block (npb)
> > > > > +	 * won't exceed maxrecs even for the blocks that get an extra record,
> > > > > +	 * since desired_npb could be maxrecs, and in the previous step we
> > > > > +	 * rounded the block count down.
> > > > > +	 */
> > > > > +	npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > > > +	if (npb > maxnr || (npb == maxnr && *blocks_with_extra > 0)) {
> > > > > +		(*blocks)++;
> > > > > +		npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > > > +	}
> > > > > +
> > > > > +	*avg_per_block = min_t(uint64_t, npb, nr_this_level);
> > > > > +
> > > > > +	trace_xfs_btree_bload_level_geometry(cur, level, nr_this_level,
> > > > > +			*avg_per_block, desired_npb, *blocks,
> > > > > +			*blocks_with_extra);
> > > > > +}
> > > > > +
> > > > > +/*
> > > > > + * Ensure a slack value is appropriate for the btree.
> > > > > + *
> > > > > + * If the slack value is negative, set slack so that we fill the block to
> > > > > + * halfway between minrecs and maxrecs.  Make sure the slack is never so large
> > > > > + * that we can underflow minrecs.
> > > > > + */
> > > > > +static void
> > > > > +xfs_btree_bload_ensure_slack(
> > > > > +	struct xfs_btree_cur	*cur,
> > > > > +	int			*slack,
> > > > > +	int			level)
> > > > > +{
> > > > > +	int			maxr;
> > > > > +	int			minr;
> > > > > +
> > > > > +	/*
> > > > > +	 * We only care about slack for btree blocks, so set the btree nlevels
> > > > > +	 * to 3 so that level 0 is a leaf block and level 1 is a node block.
> > > > > +	 * Avoid straying into inode roots, since we don't do slack there.
> > > > > +	 */
> > > > > +	cur->bc_nlevels = 3;
> > > > > +	maxr = cur->bc_ops->get_maxrecs(cur, level);
> > > > > +	minr = cur->bc_ops->get_minrecs(cur, level);
> > > > > +
> > > > > +	/*
> > > > > +	 * If slack is negative, automatically set slack so that we load the
> > > > > +	 * btree block approximately halfway between minrecs and maxrecs.
> > > > > +	 * Generally, this will net us 75% loading.
> > > > > +	 */
> > > > > +	if (*slack < 0)
> > > > > +		*slack = maxr - ((maxr + minr) >> 1);
> > > > > +
> > > > > +	*slack = min(*slack, maxr - minr);
> > > > > +}
> > > > > +
> > > > > +/*
> > > > > + * Prepare a btree cursor for a bulk load operation by computing the geometry
> > > > > + * fields in @bbl.  Caller must ensure that the btree cursor is a staging
> > > > > + * cursor.  This function can be called multiple times.
> > > > > + */
> > > > > +int
> > > > > +xfs_btree_bload_compute_geometry(
> > > > > +	struct xfs_btree_cur	*cur,
> > > > > +	struct xfs_btree_bload	*bbl,
> > > > > +	uint64_t		nr_records)
> > > > > +{
> > > > > +	uint64_t		nr_blocks = 0;
> > > > > +	uint64_t		nr_this_level;
> > > > > +
> > > > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > > > +
> > > > > +	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
> > > > > +	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
> > > > > +
> > > > > +	bbl->nr_records = nr_this_level = nr_records;
> > > > > +	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
> > > > > +		uint64_t	level_blocks;
> > > > > +		uint64_t	dontcare64;
> > > > > +		unsigned int	level = cur->bc_nlevels - 1;
> > > > > +		unsigned int	avg_per_block;
> > > > > +
> > > > > +		/*
> > > > > +		 * If all the things we want to store at this level would fit
> > > > > +		 * in a single root block, then we have our btree root and are
> > > > > +		 * done.  Note that bmap btrees do not allow records in the
> > > > > +		 * root.
> > > > > +		 */
> > > > > +		if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) || level != 0) {
> > > > > +			xfs_btree_bload_level_geometry(cur, bbl, level,
> > > > > +					nr_this_level, &avg_per_block,
> > > > > +					&level_blocks, &dontcare64);
> > > > > +			if (nr_this_level <= avg_per_block) {
> > > > > +				nr_blocks++;
> > > > > +				break;
> > > > > +			}
> > > > > +		}
> > > > > +
> > > > > +		/*
> > > > > +		 * Otherwise, we have to store all the records for this level
> > > > > +		 * in blocks and therefore need another level of btree to point
> > > > > +		 * to those blocks.  Increase the number of levels and
> > > > > +		 * recompute the number of records we can store at this level
> > > > > +		 * because that can change depending on whether or not a level
> > > > > +		 * is the root level.
> > > > > +		 */
> > > > > +		cur->bc_nlevels++;
> > > > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > > +				&avg_per_block, &level_blocks, &dontcare64);
> > > > > +		nr_blocks += level_blocks;
> > > > > +		nr_this_level = level_blocks;
> > > > > +	}
> > > > > +
> > > > > +	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
> > > > > +		return -EOVERFLOW;
> > > > > +
> > > > > +	bbl->btree_height = cur->bc_nlevels;
> > > > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
> > > > > +		bbl->nr_blocks = nr_blocks - 1;
> > > > > +	else
> > > > > +		bbl->nr_blocks = nr_blocks;
> > > > > +	return 0;
> > > > > +}
> > > > > +
> > > > > +/*
> > > > > + * Bulk load a btree.
> > > > > + *
> > > > > + * Load @bbl->nr_records quantity of records into a btree using the supplied
> > > > > + * empty and staging btree cursor @cur and a @bbl that has been filled out by
> > > > > + * the xfs_btree_bload_compute_geometry function.
> > > > > + *
> > > > > + * The @bbl->get_data function must populate the cursor's bc_rec every time it
> > > > > + * is called.  The @bbl->alloc_block function will be used to allocate new
> > > > > + * btree blocks.  @priv is passed to both functions.
> > > > > + *
> > > > > + * Caller must ensure that @cur is a staging cursor.  Any existing btree rooted
> > > > > + * in the fakeroot will be lost, so do not call this function twice.
> > > > > + */
> > > > > +int
> > > > > +xfs_btree_bload(
> > > > > +	struct xfs_btree_cur		*cur,
> > > > > +	struct xfs_btree_bload		*bbl,
> > > > > +	void				*priv)
> > > > > +{
> > > > > +	union xfs_btree_ptr		child_ptr;
> > > > > +	union xfs_btree_ptr		ptr;
> > > > > +	struct xfs_buf			*bp = NULL;
> > > > > +	struct xfs_btree_block		*block = NULL;
> > > > > +	uint64_t			nr_this_level = bbl->nr_records;
> > > > > +	uint64_t			blocks;
> > > > > +	uint64_t			i;
> > > > > +	uint64_t			blocks_with_extra;
> > > > > +	uint64_t			total_blocks = 0;
> > > > > +	unsigned int			avg_per_block;
> > > > > +	unsigned int			level = 0;
> > > > > +	int				ret;
> > > > > +
> > > > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > > > +
> > > > > +	cur->bc_nlevels = bbl->btree_height;
> > > > > +	xfs_btree_set_ptr_null(cur, &child_ptr);
> > > > > +	xfs_btree_set_ptr_null(cur, &ptr);
> > > > > +
> > > > > +	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > > +			&avg_per_block, &blocks, &blocks_with_extra);
> > > > > +
> > > > > +	/* Load each leaf block. */
> > > > > +	for (i = 0; i < blocks; i++) {
> > > > > +		unsigned int		nr_this_block = avg_per_block;
> > > > > +
> > > > > +		if (i < blocks_with_extra)
> > > > > +			nr_this_block++;
> > > > > +
> > > > > +		ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > > > +				nr_this_block, &ptr, &bp, &block, priv);
> > > > > +		if (ret)
> > > > > +			return ret;
> > > > > +
> > > > > +		trace_xfs_btree_bload_block(cur, level, i, blocks, &ptr,
> > > > > +				nr_this_block);
> > > > > +
> > > > > +		ret = xfs_btree_bload_leaf(cur, nr_this_block, bbl->get_data,
> > > > > +				block, priv);
> > > > > +		if (ret)
> > > > > +			goto out;
> > > > > +
> > > > > +		/* Record the leftmost pointer to start the next level. */
> > > > > +		if (i == 0)
> > > > > +			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
> > > > > +	}
> > > > > +	total_blocks += blocks;
> > > > > +	xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > > > +
> > > > > +	/* Populate the internal btree nodes. */
> > > > > +	for (level = 1; level < cur->bc_nlevels; level++) {
> > > > > +		union xfs_btree_ptr	first_ptr;
> > > > > +
> > > > > +		nr_this_level = blocks;
> > > > > +		block = NULL;
> > > > > +		xfs_btree_set_ptr_null(cur, &ptr);
> > > > > +
> > > > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > > +				&avg_per_block, &blocks, &blocks_with_extra);
> > > > > +
> > > > > +		/* Load each node block. */
> > > > > +		for (i = 0; i < blocks; i++) {
> > > > > +			unsigned int	nr_this_block = avg_per_block;
> > > > > +
> > > > > +			if (i < blocks_with_extra)
> > > > > +				nr_this_block++;
> > > > > +
> > > > > +			ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > > > +					nr_this_block, &ptr, &bp, &block,
> > > > > +					priv);
> > > > > +			if (ret)
> > > > > +				return ret;
> > > > > +
> > > > > +			trace_xfs_btree_bload_block(cur, level, i, blocks,
> > > > > +					&ptr, nr_this_block);
> > > > > +
> > > > > +			ret = xfs_btree_bload_node(cur, nr_this_block,
> > > > > +					&child_ptr, block);
> > > > > +			if (ret)
> > > > > +				goto out;
> > > > > +
> > > > > +			/*
> > > > > +			 * Record the leftmost pointer to start the next level.
> > > > > +			 */
> > > > > +			if (i == 0)
> > > > > +				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
> > > > > +		}
> > > > > +		total_blocks += blocks;
> > > > > +		xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > > > +		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
> > > > > +	}
> > > > > +
> > > > > +	/* Initialize the new root. */
> > > > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
> > > > > +		ASSERT(xfs_btree_ptr_is_null(cur, &ptr));
> > > > > +		cur->bc_private.b.ifake->if_levels = cur->bc_nlevels;
> > > > > +		cur->bc_private.b.ifake->if_blocks = total_blocks - 1;
> > > > > +	} else {
> > > > > +		cur->bc_private.a.afake->af_root = be32_to_cpu(ptr.s);
> > > > > +		cur->bc_private.a.afake->af_levels = cur->bc_nlevels;
> > > > > +		cur->bc_private.a.afake->af_blocks = total_blocks;
> > > > > +	}
> > > > > +out:
> > > > > +	if (bp)
> > > > > +		xfs_trans_brelse(cur->bc_tp, bp);
> > > > > +	return ret;
> > > > > +}
> > > > > diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
> > > > > index a17becb72ab8..5c6992a04ea2 100644
> > > > > --- a/fs/xfs/libxfs/xfs_btree.h
> > > > > +++ b/fs/xfs/libxfs/xfs_btree.h
> > > > > @@ -582,4 +582,47 @@ void xfs_btree_stage_ifakeroot(struct xfs_btree_cur *cur,
> > > > >  void xfs_btree_commit_ifakeroot(struct xfs_btree_cur *cur, int whichfork,
> > > > >  		const struct xfs_btree_ops *ops);
> > > > >  
> > > > > +typedef int (*xfs_btree_bload_get_fn)(struct xfs_btree_cur *cur, void *priv);
> > > > > +typedef int (*xfs_btree_bload_alloc_block_fn)(struct xfs_btree_cur *cur,
> > > > > +		union xfs_btree_ptr *ptr, void *priv);
> > > > > +typedef size_t (*xfs_btree_bload_iroot_size_fn)(struct xfs_btree_cur *cur,
> > > > > +		unsigned int nr_this_level, void *priv);
> > > > > +
> > > > > +/* Bulk loading of staged btrees. */
> > > > > +struct xfs_btree_bload {
> > > > > +	/* Function to store a record in the cursor. */
> > > > > +	xfs_btree_bload_get_fn		get_data;
> > > > > +
> > > > > +	/* Function to allocate a block for the btree. */
> > > > > +	xfs_btree_bload_alloc_block_fn	alloc_block;
> > > > > +
> > > > > +	/* Function to compute the size of the in-core btree root block. */
> > > > > +	xfs_btree_bload_iroot_size_fn	iroot_size;
> > > > > +
> > > > > +	/* Number of records the caller wants to store. */
> > > > > +	uint64_t			nr_records;
> > > > > +
> > > > > +	/* Number of btree blocks needed to store those records. */
> > > > > +	uint64_t			nr_blocks;
> > > > > +
> > > > > +	/*
> > > > > +	 * Number of free records to leave in each leaf block.  If this (or
> > > > > +	 * any of the slack values) are negative, this will be computed to
> > > > > +	 * be halfway between maxrecs and minrecs.  This typically leaves the
> > > > > +	 * block 75% full.
> > > > > +	 */
> > > > > +	int				leaf_slack;
> > > > > +
> > > > > +	/* Number of free keyptrs to leave in each node block. */
> > > > > +	int				node_slack;
> > > > > +
> > > > > +	/* Computed btree height. */
> > > > > +	unsigned int			btree_height;
> > > > > +};
> > > > > +
> > > > > +int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
> > > > > +		struct xfs_btree_bload *bbl, uint64_t nr_records);
> > > > > +int xfs_btree_bload(struct xfs_btree_cur *cur, struct xfs_btree_bload *bbl,
> > > > > +		void *priv);
> > > > > +
> > > > >  #endif	/* __XFS_BTREE_H__ */
> > > > > diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
> > > > > index bc85b89f88ca..9b5e58a92381 100644
> > > > > --- a/fs/xfs/xfs_trace.c
> > > > > +++ b/fs/xfs/xfs_trace.c
> > > > > @@ -6,6 +6,7 @@
> > > > >  #include "xfs.h"
> > > > >  #include "xfs_fs.h"
> > > > >  #include "xfs_shared.h"
> > > > > +#include "xfs_bit.h"
> > > > >  #include "xfs_format.h"
> > > > >  #include "xfs_log_format.h"
> > > > >  #include "xfs_trans_resv.h"
> > > > > diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
> > > > > index a78055521fcd..6d7ba64b7a0f 100644
> > > > > --- a/fs/xfs/xfs_trace.h
> > > > > +++ b/fs/xfs/xfs_trace.h
> > > > > @@ -35,6 +35,7 @@ struct xfs_icreate_log;
> > > > >  struct xfs_owner_info;
> > > > >  struct xfs_trans_res;
> > > > >  struct xfs_inobt_rec_incore;
> > > > > +union xfs_btree_ptr;
> > > > >  
> > > > >  DECLARE_EVENT_CLASS(xfs_attr_list_class,
> > > > >  	TP_PROTO(struct xfs_attr_list_context *ctx),
> > > > > @@ -3670,6 +3671,90 @@ TRACE_EVENT(xfs_btree_commit_ifakeroot,
> > > > >  		  __entry->blocks)
> > > > >  )
> > > > >  
> > > > > +TRACE_EVENT(xfs_btree_bload_level_geometry,
> > > > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > > > +		 uint64_t nr_this_level, unsigned int nr_per_block,
> > > > > +		 unsigned int desired_npb, uint64_t blocks,
> > > > > +		 uint64_t blocks_with_extra),
> > > > > +	TP_ARGS(cur, level, nr_this_level, nr_per_block, desired_npb, blocks,
> > > > > +		blocks_with_extra),
> > > > > +	TP_STRUCT__entry(
> > > > > +		__field(dev_t, dev)
> > > > > +		__field(xfs_btnum_t, btnum)
> > > > > +		__field(unsigned int, level)
> > > > > +		__field(unsigned int, nlevels)
> > > > > +		__field(uint64_t, nr_this_level)
> > > > > +		__field(unsigned int, nr_per_block)
> > > > > +		__field(unsigned int, desired_npb)
> > > > > +		__field(unsigned long long, blocks)
> > > > > +		__field(unsigned long long, blocks_with_extra)
> > > > > +	),
> > > > > +	TP_fast_assign(
> > > > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > > > +		__entry->btnum = cur->bc_btnum;
> > > > > +		__entry->level = level;
> > > > > +		__entry->nlevels = cur->bc_nlevels;
> > > > > +		__entry->nr_this_level = nr_this_level;
> > > > > +		__entry->nr_per_block = nr_per_block;
> > > > > +		__entry->desired_npb = desired_npb;
> > > > > +		__entry->blocks = blocks;
> > > > > +		__entry->blocks_with_extra = blocks_with_extra;
> > > > > +	),
> > > > > +	TP_printk("dev %d:%d btree %s level %u/%u nr_this_level %llu nr_per_block %u desired_npb %u blocks %llu blocks_with_extra %llu",
> > > > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > > > +		  __entry->level,
> > > > > +		  __entry->nlevels,
> > > > > +		  __entry->nr_this_level,
> > > > > +		  __entry->nr_per_block,
> > > > > +		  __entry->desired_npb,
> > > > > +		  __entry->blocks,
> > > > > +		  __entry->blocks_with_extra)
> > > > > +)
> > > > > +
> > > > > +TRACE_EVENT(xfs_btree_bload_block,
> > > > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > > > +		 uint64_t block_idx, uint64_t nr_blocks,
> > > > > +		 union xfs_btree_ptr *ptr, unsigned int nr_records),
> > > > > +	TP_ARGS(cur, level, block_idx, nr_blocks, ptr, nr_records),
> > > > > +	TP_STRUCT__entry(
> > > > > +		__field(dev_t, dev)
> > > > > +		__field(xfs_btnum_t, btnum)
> > > > > +		__field(unsigned int, level)
> > > > > +		__field(unsigned long long, block_idx)
> > > > > +		__field(unsigned long long, nr_blocks)
> > > > > +		__field(xfs_agnumber_t, agno)
> > > > > +		__field(xfs_agblock_t, agbno)
> > > > > +		__field(unsigned int, nr_records)
> > > > > +	),
> > > > > +	TP_fast_assign(
> > > > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > > > +		__entry->btnum = cur->bc_btnum;
> > > > > +		__entry->level = level;
> > > > > +		__entry->block_idx = block_idx;
> > > > > +		__entry->nr_blocks = nr_blocks;
> > > > > +		if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
> > > > > +			xfs_fsblock_t	fsb = be64_to_cpu(ptr->l);
> > > > > +
> > > > > +			__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsb);
> > > > > +			__entry->agbno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsb);
> > > > > +		} else {
> > > > > +			__entry->agno = cur->bc_private.a.agno;
> > > > > +			__entry->agbno = be32_to_cpu(ptr->s);
> > > > > +		}
> > > > > +		__entry->nr_records = nr_records;
> > > > > +	),
> > > > > +	TP_printk("dev %d:%d btree %s level %u block %llu/%llu fsb (%u/%u) recs %u",
> > > > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > > > +		  __entry->level,
> > > > > +		  __entry->block_idx,
> > > > > +		  __entry->nr_blocks,
> > > > > +		  __entry->agno,
> > > > > +		  __entry->agbno,
> > > > > +		  __entry->nr_records)
> > > > > +)
> > > > > +
> > > > >  #endif /* _TRACE_XFS_H */
> > > > >  
> > > > >  #undef TRACE_INCLUDE_PATH
> > > > >

Darrick J. Wong Oct. 17, 2019, 7:06 p.m. UTC | #6

On Thu, Oct 17, 2019 at 05:32:59AM -0400, Brian Foster wrote:
> On Wed, Oct 16, 2019 at 05:40:18PM -0700, Darrick J. Wong wrote:
> > On Wed, Oct 16, 2019 at 05:07:31PM -0400, Brian Foster wrote:
> > > On Wed, Oct 16, 2019 at 11:15:02AM -0700, Darrick J. Wong wrote:
> > > > On Wed, Oct 16, 2019 at 11:26:48AM -0400, Brian Foster wrote:
> > > > > On Wed, Oct 09, 2019 at 09:48:18AM -0700, Darrick J. Wong wrote:
> > > > > > From: Darrick J. Wong <[e-mail]>
> > > > > > 
> > > > > > Add a new btree function that enables us to bulk load a btree cursor.
> > > > > > This will be used by the upcoming online repair patches to generate new
> > > > > > btrees.
> > > > > > 
> > > > > > Signed-off-by: Darrick J. Wong <[e-mail]>
> > > > > > ---
> > > > > >  fs/xfs/libxfs/xfs_btree.c |  566 +++++++++++++++++++++++++++++++++++++++++++++
> > > > > >  fs/xfs/libxfs/xfs_btree.h |   43 +++
> > > > > >  fs/xfs/xfs_trace.c        |    1 
> > > > > >  fs/xfs/xfs_trace.h        |   85 +++++++
> > > > > >  4 files changed, 694 insertions(+), 1 deletion(-)
> > > > > > 
> > > > > > 
> > > > > > diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
> > > > > > index 4b06d5d86834..17b0fdb87729 100644
> > > > > > --- a/fs/xfs/libxfs/xfs_btree.c
> > > > > > +++ b/fs/xfs/libxfs/xfs_btree.c
> > > > > ...
> > > > > > @@ -5104,3 +5104,567 @@ xfs_btree_commit_ifakeroot(
> > > > > >  	cur->bc_ops = ops;
> > > > > >  	cur->bc_flags &= ~XFS_BTREE_STAGING;
> > > > > >  }
> > > > > > +
> > > > > > +/*
> > > > > > + * Bulk Loading of Staged Btrees
> > > > > > + * =============================
> > > > > > + *
> > > > > > + * This interface is used with a staged btree cursor to create a totally new
> > > > > > + * btree with a large number of records (i.e. more than what would fit in a
> > > > > > + * single block).  When the creation is complete, the new root can be linked
> > > > > > + * atomically into the filesystem by committing the staged cursor.
> > > > > > + *
> > > > 
> > > > [paraphrasing a conversation we had on irc]
> > > > 
> > > > > Thanks for the documentation. So what is the purpose behind the whole
> > > > > bulk loading thing as opposed to something like faking up an AG
> > > > > structure (i.e. AGF) somewhere and using the existing cursor mechanisms
> > > > > (or something closer to it) to copy records from one place to another?
> > > > > Is it purely a performance/efficiency tradeoff? Bulk block allocation
> > > > > issues? Transactional/atomicity issues? All (or none :P) of the above?
> > > > 
> > > > Prior to the v20, the online repair series created a new btree root,
> > > > committed that into wherever the root lived, and inserted records one by
> > > > one into the btree.  There were quite a few drawbacks to this method:
> > > > 
> > > > 1. Inserting records one at a time can involve walking up the tree to
> > > > update node block pointers, which isn't terribly efficient if we're
> > > > likely going to rewrite the pointers (and relogging nodes) several more
> > > > times.
> > > > 
> > > > 2. Inserting records one at a time tends to leave a lot of half-empty
> > > > btree blocks because when one block fills up we split it and push half
> > > > the records to the new block.  It would be nice not to explode the size
> > > > of the btrees, and it would be particularly useful if we could control
> > > > the load factor of the new btree precisely.
> > > > 
> > > 
> > > Interesting... this is a trait the traditional btree update paths share
> > > though, right?
> > 
> > Right.  It's similar to the behavior Dave was seeing a couple of weeks
> > ago with Zorro's stress testing of the incore extent cache.
> > 
> > > > 3. The rebuild wasn't atomic, since we were replacing the root prior to
> > > > the insert loop.  If we crashed midway through a rebuild we'd end up
> > > > with a garbage btree and no indication that it was incorrect.  That's
> > > > how the fakeroot code got started.
> > > > 
> > > 
> > > Indeed, though this seems more related to the anchoring (i.e. fake root)
> > > approach than bulk vs. iterative construction.
> > 
> > Correct.
> > 
> > > > 4. In a previous version of the repair series I tried to batch as many
> > > > insert operations into a single transaction as possible, but my
> > > > transaction reservation fullness estimation function didn't work
> > > > reliably (particularly when things got really fragmented), so I backed
> > > > off to rolling after /every/ insertion.  That works well enough, but at
> > > > a cost of a lot of transaction rolling, which means that repairs plod
> > > > along very slowly.
> > > > 
> > > > 5. Performing an insert loop means that the btree blocks are allocated
> > > > one at a time as the btree expands.  This is suboptimal since we can
> > > > calculate the exact size of the new btree prior to building it, which
> > > > gives us the opportunity to recreate the index in a set of contiguous
> > > > blocks instead of scattering them.
> > > > 
> > > 
> > > Yep, FWIW it sounds like most of these tradeoffs are around
> > > performance/efficiency. 
> > 
> > <nod>
> > 
> > > > 6. If we crash midway through a rebuild, XFS neither cleaned up the mess
> > > > nor informed the administrator that it was necessary to re-run xfs_scrub
> > > > or xfs_repair to clean up the lost blocks.  Obviously, automatic cleanup
> > > > is a far better solution.
> > > > 
> > > 
> > > Similar to above, I think this kind of depends more on how/where to
> > > anchor an in-progress tree as opposed to what level records are copied
> > > at.
> > 
> > <nod> The six points are indeed the overall list of complaints about the
> > v19 code. :)
> > 
> > > > The first thing I decided to solve was the lack of atomicity.
> > > > 
> > > > For AG-rooted btrees, I thought about creating a fake xfs_buf for an AG
> > > > header buffer and extracting the root/level values after construction
> > > > completes.  That's possible, but it's risky because the fake buffer
> > > > could get logged and if the sector number matches the actual header
> > > > then it introduces buffer cache aliasing issues.
> > > > 
> > > > For inode-rooted btrees, one could create a fake xfs_inode with the same
> > > > i_ino as the target.  That presents the same aliasing issues as the fake
> > > > xfs_buf above.  A different strategy would be to allocate an unlinked
> > > > inode and then use the bmbt owner change (a.k.a. extent swap) to move
> > > > the mappings over.  That would work, though it has two large drawbacks:
> > > > (a) a lot of additional complexity around allocating and freeing the
> > > > temporary inode; and (b) future inode-rooted btrees such as the realtime
> > > > rmap btree would also have to implement an owner-change operation.
> > > > 
> > > 
> > > I was wondering more along the lines of having an actual anchor
> > > somewhere. E.g., think of it as a temporary/inaccessible location of a
> > > legitimate on-disk structure as opposed to a fake object in memory
> > > somewhere. A hidden/internal repair inode or some such, perhaps. I'm
> > > sure there's new code/complexity that would come around with that, but I
> > > think that's going to be unavoidable to some degree for an online repair
> > > mechanism. ;)
> > 
> > <nod> So far I /think/ I've managed to keep to an absolute minimum the
> > amount of metadata that gets written to disk prior to the commit.  I
> > haven't reread the series with an eye for how v20 is going to come up
> > short though. :)
> > 
> 
> This metadata has to be written out one way or another though. IOW, it
> seems the advantage to holding off is mostly for the case where we crash
> or something mid reconstruction (which should be the uncommon case :P).

Right.

> I could see how there might be some benefits to using contiguous blocks
> in a btree and batching them out as such, but conversely I'm also
> curious about how that might behave with certain scenarios like really
> large btrees that take a while to reconstruct (e.g. due to read
> contention or slow storage) or where contiguous blocks aren't available,
> etc.

You mean like a debug knob or something to force the allocation to
scatter blocks everywhere?  I do some regular testing of large(r) btree
reconstruction with a metadump of /home that's been mdrestored to a
thinp device.

(Or I guess building an xfstests that sets up a slow dm device...?)

> > I may very well have to revisit the hidden/internal repair inode concept
> > whenever I start working on rebuilding directories and xattrs since I
> > can't see any other way of atomically rebuilding those.  But that's
> > very very far out still.
> > 
> 
> Ok.
> 
> > > Note that this is all just handwaving on my part and still without full
> > > context as to how things are currently anchored, made atomic, etc. I'm
> > > primarily trying to understand the design reasoning based on the high
> > > level description.
> > 
> > <nod>
> > 
> > > > To fix (3), I thought it wise to have explicit fakeroot structures to
> > > > maintain a clean separation between what we're building and the rest of
> > > > the filesystem.  This also means that there's nothing on disk to clean
> > > > up if we fail at any point before we're ready to commit the new btree.
> > > > 
> > > 
> > > Hmm.. so this approach facilites a tree reconstruction in a single open
> > > transaction? If so, I suppose I could see some functional advantages to
> > > that.
> > 
> > Correct.
> > 
> > > > Then Dave (I think?) suggested that I  use EFIs strategically to
> > > > schedule freeing of the new btree blocks (the root commit transaction
> > > > would log EFDs to cancel them) and to schedule freeing of the old
> > > > blocks.  That solves (6), though the EFI wrangling doesn't happen for
> > > > another couple of series after this one.
> > > > 
> > > 
> > > Hm, Ok... so new btree block allocation(s?) in the same transaction as
> > > an EFI, to be processed on recovery if we crash, otherwise cancelled
> > > with an EFD on construction completion..?
> > 
> > Correct.  In the end, the transaction sequence looks like:
> > 
> > T[1]: Allocate an extent, log metadata updates to reflect that, log EFI
> > for the extent.
> > 
> > <repeat until we've allocated as many blocks as we need>
> > 
> 
> If I follow correctly, I take it "repeat" here means we're rolling the
> transaction to reallocate, and relog all previous EFIs along the way,
> until we've acquired sufficient blocks to start populating the tree.

Right, though I don't explicitly relog the previous EFIs.

> > T[N]: Attach ordered buffers for the new btree's blocks.  Log the root
> > change.  Log EFDs for all the EFIs logged in T[1..N-1].  Log EFIs for
> > all the old btree blocks that we could find.
> > 
> 
> So here we're on/after the last allocation transaction, have a bunch of
> pending EFIs for recovery purposes, and start populating the btree
> blocks...
> 
> AFAICT the log tail is pinned by these EFIs during population, which is
> of non-deterministic time for aforementioned reasons. If so, this
> _sounds_ like it risks a similar log deadlock vector to the one we
> currently have for quotaoff, though I could easily be missing something
> from the high level description.

Yes, the EFIs pin the log and put us at risk for a quotaoff style log
deadlock.  This was particularly risky with the insert_rec loop since
repair itself was contributing to log traffic.  I've wondered what would
happen if we could relog intent items the same way we do with inode
cores to avoid pinning the tail?

I guess it's /not/ that similar since the log itself could relog the
pinned intent items since they're not supposed to change once they've
been logged the first time.

> Also, what happens if we complete the population, roll this transaction,
> unpin the ordered buffers and crash before the entire set of ordered
> buffers is written back? We currently use ordered buffers with
> operations described by logical intents. For example, an icreate item
> goes into the log with an ordered buffer such that if we crash before
> buffer writeback completes and allows the tail to move, recovery knows
> how to turn the active icreate log item back into a valid inode chunk.
> Is there something similar going on here?

Ooooh, yes.  Paraphrasing (and expanding upon) what discussed on IRC
earlier, I forgot that ordered buffers are written out by the AIL, which
means that xfs_btree_bload needs to force the new blocks to be written
to disk before logging and committing the new root.

Ok, so I guess we need xfs_btree_bload to delwri_queue each new block's
buffer into a list and then delwri_submit the list before returning.  If
we hit an error we'll cancel the transaction, if we succeed then we move
on to commit the staging cursor, etc.

Hmm, thinking about it further, if the delwri_submit fails we could
commit the transaction and return from repair without committing the
root.  The EFIs will trigger, which frees the blocks we allocated, and
the fs ends in the same place it was before the repair started.

Though that's an optimization; for now we'll just cancel the
transaction and bail out.

> > <roll transaction to write the ordered buffers and commit>
> > 
> > T[N+1]: Free an extent to finish the first EFI logged in the previous step.
> > 
> > <repeat until we've processed everything from the second wave of EFIs>
> > 
> 
> And I take repeat here means essentially rolling and completing the
> "real" EFIs for the old btree we've swapped out.

Yes.

> FWIW, it seems to me that this altogether might be an opportunity for a
> custom logical intent to cover reconstruction sequences. 

How do you mean?  A new "btree reconstruction intent" log item?  Or
maybe more generally, a "repair intent" log item?  Hmm, that would force
a new log_incompat flag, but I'll think about that.  So far my thinking
has been that a failed online repair should be followed up with an
xfs_repair run, just in case the repair failed due to software bug.  But
maybe recovery of the new intent type would log a message and return
error to force xfs_repair, and when we're more confident we can change
it to call into xfs_scrub_metadata().

> > Call xfs_trans_commit and we're done.
> > 
> > > > He also suggested using ordered buffers to write out the new btree
> > > > blocks along with whatever logging was necessary to commit the new
> > > > btree.  It then occurred to me that xfs_repair open-codes the process of
> > > > calculating the geometry of a new btree, allocating all the blocks at
> > > > once, and writing out full btree blocks.  Somewhat annoyingly, it
> > > > features nearly the same (open-)code for all four AG btree types, which
> > > > is less maintainable than it could be.
> > > > 
> > > > I read through all four versions and used it to write the generic btree
> > > > bulk loading code.  For scrub I hooked that up to the "staged btree with
> > > > a fake root" stuff I'd already written, which solves (1), (2), (4), and
> > > > (5).
> > > > 
> > > > For xfsprogs[1], I deleted a few thousand lines of code from xfs_repair.
> > > > True, we don't reuse existing common code, but we at least get to share
> > > > new common btree code.
> > > > 
> > > 
> > > Yeah, the xfsprogs work certainly makes sense. Part of the reason I ask
> > > about this is the tradeoff of having multiple avenues to construct a
> > > tree in the kernel codebase.
> > 
> > <nod>
> > 
> > > > > This is my first pass through this so I'm mostly looking at big picture
> > > > > until I get to a point to see how these bits are used. The mechanism
> > > > > itself seems reasonable in principle, but the reason I ask is it also
> > > > > seems like there's inherent value in using more of same infrastructure
> > > > > to reconstruct a tree that we use to create one in the first place. We
> > > > > also already have primitives for things like fork swapping via the
> > > > > extent swap mechanism, etc.
> > > > 
> > > > "bfoster: I guess it would be nice to see that kind of make it work ->
> > > > make it fast evolution in tree"
> > > > 
> > > > For a while I did maintain the introduction of the bulk loading code as
> > > > separate patches against the v19 repair code, but unfortunately I
> > > > smushed them down before sending v20 to reduce the patch count, and
> > > > because I didn't want to argue with everyone over the semi-working code
> > > > that would then be replaced in the very next patch.
> > > > 
> > > 
> > > That's not quite what I meant... The approach you've taken makes sense
> > > to me for an implementation presented in a single series. I was more
> > > thinking that at the point where it was determined the implementation
> > > was going to change so drastically, after so many iterations it might
> > > have been useful to see the v19 approach merged in an experimental form
> > 
> > I would have liked to see the online repair stuff merged in experimental
> > form too so I can reduce the size of my patch queue, but oh well. :)
> > 
> 
> Oh well, indeed. :P

Well, admittedly, it's /still/ a lot of half-baked code. :D

> > The silver lining to these lengthy reworks and slow review is that I can
> > come back and do a fresh self-review after a month and straighten out
> > the ugly parts as time goes by.  Unfortunately, that doesn't leave much
> > of a paper trail or obvious evidence of development history.
> > 
> 
> Exactly.
> 
> > Eight months ago it occurred to me that perhaps there is some value in
> > retaining *some* periodic development history of this, so I've been
> > adding dated tags to my integration repo[1] based on my development
> > branch names, so I guess you could actually clone the git repo and git
> > diff from one tag to another.  In general I'll generate a new pile of
> > tags just before patchbombing.
> > 
> 
> More importantly, I think we should try to follow that approach if
> possible from here on, even if there are open/unresolved issues with the
> currently proposed approach. Online repair overall is complex enough
> that it would probably be impossible to get a fully functional online
> repair done in a single series (and we're clearly not taking that
> approach anyways), making sure that the various common underpinnings
> work for all of the various types of structures, etc. Just my .02
> though..

<nod>  I'll keep posting the tags and whatnot to my private repo and
leave them there.  Anyone else pushing large patchsets could do
likewise.

--D

> Brian
> 
> > (Granted 'repair-part-one' has been split into smaller parts now...)
> > 
> > > and then reworked upstream from there. Now that the new approach is
> > > implemented, I agree it's probably not worth reinserting the old
> > > approach at this point just to switch it out.
> > 
> > <nod>
> > 
> > > Thanks for the breakdown...
> > 
> > No problem, thanks for reading! :)
> > 
> > --D
> > 
> > [1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux.git
> > 
> > > Brian
> > > 
> > > > I could split them back out, though at a cost of having to reintroduce a
> > > > lot of hairy code in the bnobt/cntbt rebuild function to seed the free
> > > > new space btree root in order to make sure that the btree block
> > > > allocation code works properly, along with auditing the allocation paths
> > > > to make sure they don't use the old AGF or encounter other subtleties.
> > > > 
> > > > It'd be a lot of work considering that the v20 reconstruction code is
> > > > /much/ simpler than v19's was.  I also restructured the repair functions
> > > > to allocate one large context structure at the beginning instead of the
> > > > piecemeal way it was done onstack in v19 because stack usage was growing
> > > > close to 1k in some cases.
> > > > 
> > > > --D
> > > > 
> > > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/log/?h=repair-bulk-load
> > > > 
> > > > > 
> > > > > Brian
> > > > > 
> > > > > > + * The first step for the caller is to construct a fake btree root structure
> > > > > > + * and a staged btree cursor.  A staging cursor contains all the geometry
> > > > > > + * information for the btree type but will fail all operations that could have
> > > > > > + * side effects in the filesystem (e.g. btree shape changes).  Regular
> > > > > > + * operations will not work unless the staging cursor is committed and becomes
> > > > > > + * a regular cursor.
> > > > > > + *
> > > > > > + * For a btree rooted in an AG header, use an xbtree_afakeroot structure.
> > > > > > + * This should be initialized to zero.  For a btree rooted in an inode fork,
> > > > > > + * use an xbtree_ifakeroot structure.  @if_fork_size field should be set to
> > > > > > + * the number of bytes available to the fork in the inode; @if_fork should
> > > > > > + * point to a freshly allocated xfs_inode_fork; and @if_format should be set
> > > > > > + * to the appropriate fork type (e.g. XFS_DINODE_FMT_BTREE).
> > > > > > + *
> > > > > > + * The next step for the caller is to initialize a struct xfs_btree_bload
> > > > > > + * context.  The @nr_records field is the number of records that are to be
> > > > > > + * loaded into the btree.  The @leaf_slack and @node_slack fields are the
> > > > > > + * number of records (or key/ptr) slots to leave empty in new btree blocks.
> > > > > > + * If a caller sets a slack value to -1, the slack value will be computed to
> > > > > > + * fill the block halfway between minrecs and maxrecs items per block.
> > > > > > + *
> > > > > > + * The number of items placed in each btree block is computed via the following
> > > > > > + * algorithm: For leaf levels, the number of items for the level is nr_records.
> > > > > > + * For node levels, the number of items for the level is the number of blocks
> > > > > > + * in the next lower level of the tree.  For each level, the desired number of
> > > > > > + * items per block is defined as:
> > > > > > + *
> > > > > > + * desired = max(minrecs, maxrecs - slack factor)
> > > > > > + *
> > > > > > + * The number of blocks for the level is defined to be:
> > > > > > + *
> > > > > > + * blocks = nr_items / desired
> > > > > > + *
> > > > > > + * Note this is rounded down so that the npb calculation below will never fall
> > > > > > + * below minrecs.  The number of items that will actually be loaded into each
> > > > > > + * btree block is defined as:
> > > > > > + *
> > > > > > + * npb =  nr_items / blocks
> > > > > > + *
> > > > > > + * Some of the leftmost blocks in the level will contain one extra record as
> > > > > > + * needed to handle uneven division.  If the number of records in any block
> > > > > > + * would exceed maxrecs for that level, blocks is incremented and npb is
> > > > > > + * recalculated.
> > > > > > + *
> > > > > > + * In other words, we compute the number of blocks needed to satisfy a given
> > > > > > + * loading level, then spread the items as evenly as possible.
> > > > > > + *
> > > > > > + * To complete this step, call xfs_btree_bload_compute_geometry, which uses
> > > > > > + * those settings to compute the height of the btree and the number of blocks
> > > > > > + * that will be needed to construct the btree.  These values are stored in the
> > > > > > + * @btree_height and @nr_blocks fields.
> > > > > > + *
> > > > > > + * At this point, the caller must allocate @nr_blocks blocks and save them for
> > > > > > + * later.  If space is to be allocated transactionally, the staging cursor
> > > > > > + * must be deleted before and recreated after, which is why computing the
> > > > > > + * geometry is a separate step.
> > > > > > + *
> > > > > > + * The fourth step in the bulk loading process is to set the function pointers
> > > > > > + * in the bload context structure.  @get_data will be called for each record
> > > > > > + * that will be loaded into the btree; it should set the cursor's bc_rec
> > > > > > + * field, which will be converted to on-disk format and copied into the
> > > > > > + * appropriate record slot.  @alloc_block should supply one of the blocks
> > > > > > + * allocated in the previous step.  For btrees which are rooted in an inode
> > > > > > + * fork, @iroot_size is called to compute the size of the incore btree root
> > > > > > + * block.  Call xfs_btree_bload to start constructing the btree.
> > > > > > + *
> > > > > > + * The final step is to commit the staging cursor, which logs the new btree
> > > > > > + * root and turns the btree into a regular btree cursor, and free the fake
> > > > > > + * roots.
> > > > > > + */
> > > > > > +
> > > > > > +/*
> > > > > > + * Put a btree block that we're loading onto the ordered list and release it.
> > > > > > + * The btree blocks will be written when the final transaction swapping the
> > > > > > + * btree roots is committed.
> > > > > > + */
> > > > > > +static void
> > > > > > +xfs_btree_bload_drop_buf(
> > > > > > +	struct xfs_trans	*tp,
> > > > > > +	struct xfs_buf		**bpp)
> > > > > > +{
> > > > > > +	if (*bpp == NULL)
> > > > > > +		return;
> > > > > > +
> > > > > > +	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_BTREE_BUF);
> > > > > > +	xfs_trans_ordered_buf(tp, *bpp);
> > > > > > +	xfs_trans_brelse(tp, *bpp);
> > > > > > +	*bpp = NULL;
> > > > > > +}
> > > > > > +
> > > > > > +/* Allocate and initialize one btree block for bulk loading. */
> > > > > > +STATIC int
> > > > > > +xfs_btree_bload_prep_block(
> > > > > > +	struct xfs_btree_cur		*cur,
> > > > > > +	struct xfs_btree_bload		*bbl,
> > > > > > +	unsigned int			level,
> > > > > > +	unsigned int			nr_this_block,
> > > > > > +	union xfs_btree_ptr		*ptrp,
> > > > > > +	struct xfs_buf			**bpp,
> > > > > > +	struct xfs_btree_block		**blockp,
> > > > > > +	void				*priv)
> > > > > > +{
> > > > > > +	union xfs_btree_ptr		new_ptr;
> > > > > > +	struct xfs_buf			*new_bp;
> > > > > > +	struct xfs_btree_block		*new_block;
> > > > > > +	int				ret;
> > > > > > +
> > > > > > +	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
> > > > > > +	    level == cur->bc_nlevels - 1) {
> > > > > > +		struct xfs_ifork	*ifp = cur->bc_private.b.ifake->if_fork;
> > > > > > +		size_t			new_size;
> > > > > > +
> > > > > > +		/* Allocate a new incore btree root block. */
> > > > > > +		new_size = bbl->iroot_size(cur, nr_this_block, priv);
> > > > > > +		ifp->if_broot = kmem_zalloc(new_size, 0);
> > > > > > +		ifp->if_broot_bytes = (int)new_size;
> > > > > > +		ifp->if_flags |= XFS_IFBROOT;
> > > > > > +
> > > > > > +		/* Initialize it and send it out. */
> > > > > > +		xfs_btree_init_block_int(cur->bc_mp, ifp->if_broot,
> > > > > > +				XFS_BUF_DADDR_NULL, cur->bc_btnum, level,
> > > > > > +				nr_this_block, cur->bc_private.b.ip->i_ino,
> > > > > > +				cur->bc_flags);
> > > > > > +
> > > > > > +		*bpp = NULL;
> > > > > > +		*blockp = ifp->if_broot;
> > > > > > +		xfs_btree_set_ptr_null(cur, ptrp);
> > > > > > +		return 0;
> > > > > > +	}
> > > > > > +
> > > > > > +	/* Allocate a new leaf block. */
> > > > > > +	ret = bbl->alloc_block(cur, &new_ptr, priv);
> > > > > > +	if (ret)
> > > > > > +		return ret;
> > > > > > +
> > > > > > +	ASSERT(!xfs_btree_ptr_is_null(cur, &new_ptr));
> > > > > > +
> > > > > > +	ret = xfs_btree_get_buf_block(cur, &new_ptr, &new_block, &new_bp);
> > > > > > +	if (ret)
> > > > > > +		return ret;
> > > > > > +
> > > > > > +	/* Initialize the btree block. */
> > > > > > +	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
> > > > > > +	if (*blockp)
> > > > > > +		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
> > > > > > +	xfs_btree_set_sibling(cur, new_block, ptrp, XFS_BB_LEFTSIB);
> > > > > > +	xfs_btree_set_numrecs(new_block, nr_this_block);
> > > > > > +
> > > > > > +	/* Release the old block and set the out parameters. */
> > > > > > +	xfs_btree_bload_drop_buf(cur->bc_tp, bpp);
> > > > > > +	*blockp = new_block;
> > > > > > +	*bpp = new_bp;
> > > > > > +	xfs_btree_copy_ptrs(cur, ptrp, &new_ptr, 1);
> > > > > > +	return 0;
> > > > > > +}
> > > > > > +
> > > > > > +/* Load one leaf block. */
> > > > > > +STATIC int
> > > > > > +xfs_btree_bload_leaf(
> > > > > > +	struct xfs_btree_cur		*cur,
> > > > > > +	unsigned int			recs_this_block,
> > > > > > +	xfs_btree_bload_get_fn		get_data,
> > > > > > +	struct xfs_btree_block		*block,
> > > > > > +	void				*priv)
> > > > > > +{
> > > > > > +	unsigned int			j;
> > > > > > +	int				ret;
> > > > > > +
> > > > > > +	/* Fill the leaf block with records. */
> > > > > > +	for (j = 1; j <= recs_this_block; j++) {
> > > > > > +		union xfs_btree_rec	*block_recs;
> > > > > > +
> > > > > > +		ret = get_data(cur, priv);
> > > > > > +		if (ret)
> > > > > > +			return ret;
> > > > > > +		block_recs = xfs_btree_rec_addr(cur, j, block);
> > > > > > +		cur->bc_ops->init_rec_from_cur(cur, block_recs);
> > > > > > +	}
> > > > > > +
> > > > > > +	return 0;
> > > > > > +}
> > > > > > +
> > > > > > +/* Load one node block. */
> > > > > > +STATIC int
> > > > > > +xfs_btree_bload_node(
> > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > +	unsigned int		recs_this_block,
> > > > > > +	union xfs_btree_ptr	*child_ptr,
> > > > > > +	struct xfs_btree_block	*block)
> > > > > > +{
> > > > > > +	unsigned int		j;
> > > > > > +	int			ret;
> > > > > > +
> > > > > > +	/* Fill the node block with keys and pointers. */
> > > > > > +	for (j = 1; j <= recs_this_block; j++) {
> > > > > > +		union xfs_btree_key	child_key;
> > > > > > +		union xfs_btree_ptr	*block_ptr;
> > > > > > +		union xfs_btree_key	*block_key;
> > > > > > +		struct xfs_btree_block	*child_block;
> > > > > > +		struct xfs_buf		*child_bp;
> > > > > > +
> > > > > > +		ASSERT(!xfs_btree_ptr_is_null(cur, child_ptr));
> > > > > > +
> > > > > > +		ret = xfs_btree_get_buf_block(cur, child_ptr, &child_block,
> > > > > > +				&child_bp);
> > > > > > +		if (ret)
> > > > > > +			return ret;
> > > > > > +
> > > > > > +		xfs_btree_get_keys(cur, child_block, &child_key);
> > > > > > +
> > > > > > +		block_ptr = xfs_btree_ptr_addr(cur, j, block);
> > > > > > +		xfs_btree_copy_ptrs(cur, block_ptr, child_ptr, 1);
> > > > > > +
> > > > > > +		block_key = xfs_btree_key_addr(cur, j, block);
> > > > > > +		xfs_btree_copy_keys(cur, block_key, &child_key, 1);
> > > > > > +
> > > > > > +		xfs_btree_get_sibling(cur, child_block, child_ptr,
> > > > > > +				XFS_BB_RIGHTSIB);
> > > > > > +		xfs_trans_brelse(cur->bc_tp, child_bp);
> > > > > > +	}
> > > > > > +
> > > > > > +	return 0;
> > > > > > +}
> > > > > > +
> > > > > > +/*
> > > > > > + * Compute the maximum number of records (or keyptrs) per block that we want to
> > > > > > + * install at this level in the btree.  Caller is responsible for having set
> > > > > > + * @cur->bc_private.b.forksize to the desired fork size, if appropriate.
> > > > > > + */
> > > > > > +STATIC unsigned int
> > > > > > +xfs_btree_bload_max_npb(
> > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > +	struct xfs_btree_bload	*bbl,
> > > > > > +	unsigned int		level)
> > > > > > +{
> > > > > > +	unsigned int		ret;
> > > > > > +
> > > > > > +	if (level == cur->bc_nlevels - 1 && cur->bc_ops->get_dmaxrecs)
> > > > > > +		return cur->bc_ops->get_dmaxrecs(cur, level);
> > > > > > +
> > > > > > +	ret = cur->bc_ops->get_maxrecs(cur, level);
> > > > > > +	if (level == 0)
> > > > > > +		ret -= bbl->leaf_slack;
> > > > > > +	else
> > > > > > +		ret -= bbl->node_slack;
> > > > > > +	return ret;
> > > > > > +}
> > > > > > +
> > > > > > +/*
> > > > > > + * Compute the desired number of records (or keyptrs) per block that we want to
> > > > > > + * install at this level in the btree, which must be somewhere between minrecs
> > > > > > + * and max_npb.  The caller is free to install fewer records per block.
> > > > > > + */
> > > > > > +STATIC unsigned int
> > > > > > +xfs_btree_bload_desired_npb(
> > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > +	struct xfs_btree_bload	*bbl,
> > > > > > +	unsigned int		level)
> > > > > > +{
> > > > > > +	unsigned int		npb = xfs_btree_bload_max_npb(cur, bbl, level);
> > > > > > +
> > > > > > +	/* Root blocks are not subject to minrecs rules. */
> > > > > > +	if (level == cur->bc_nlevels - 1)
> > > > > > +		return max(1U, npb);
> > > > > > +
> > > > > > +	return max_t(unsigned int, cur->bc_ops->get_minrecs(cur, level), npb);
> > > > > > +}
> > > > > > +
> > > > > > +/*
> > > > > > + * Compute the number of records to be stored in each block at this level and
> > > > > > + * the number of blocks for this level.  For leaf levels, we must populate an
> > > > > > + * empty root block even if there are no records, so we have to have at least
> > > > > > + * one block.
> > > > > > + */
> > > > > > +STATIC void
> > > > > > +xfs_btree_bload_level_geometry(
> > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > +	struct xfs_btree_bload	*bbl,
> > > > > > +	unsigned int		level,
> > > > > > +	uint64_t		nr_this_level,
> > > > > > +	unsigned int		*avg_per_block,
> > > > > > +	uint64_t		*blocks,
> > > > > > +	uint64_t		*blocks_with_extra)
> > > > > > +{
> > > > > > +	uint64_t		npb;
> > > > > > +	uint64_t		dontcare;
> > > > > > +	unsigned int		desired_npb;
> > > > > > +	unsigned int		maxnr;
> > > > > > +
> > > > > > +	maxnr = cur->bc_ops->get_maxrecs(cur, level);
> > > > > > +
> > > > > > +	/*
> > > > > > +	 * Compute the number of blocks we need to fill each block with the
> > > > > > +	 * desired number of records/keyptrs per block.  Because desired_npb
> > > > > > +	 * could be minrecs, we use regular integer division (which rounds
> > > > > > +	 * the block count down) so that in the next step the effective # of
> > > > > > +	 * items per block will never be less than desired_npb.
> > > > > > +	 */
> > > > > > +	desired_npb = xfs_btree_bload_desired_npb(cur, bbl, level);
> > > > > > +	*blocks = div64_u64_rem(nr_this_level, desired_npb, &dontcare);
> > > > > > +	*blocks = max(1ULL, *blocks);
> > > > > > +
> > > > > > +	/*
> > > > > > +	 * Compute the number of records that we will actually put in each
> > > > > > +	 * block, assuming that we want to spread the records evenly between
> > > > > > +	 * the blocks.  Take care that the effective # of items per block (npb)
> > > > > > +	 * won't exceed maxrecs even for the blocks that get an extra record,
> > > > > > +	 * since desired_npb could be maxrecs, and in the previous step we
> > > > > > +	 * rounded the block count down.
> > > > > > +	 */
> > > > > > +	npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > > > > +	if (npb > maxnr || (npb == maxnr && *blocks_with_extra > 0)) {
> > > > > > +		(*blocks)++;
> > > > > > +		npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > > > > +	}
> > > > > > +
> > > > > > +	*avg_per_block = min_t(uint64_t, npb, nr_this_level);
> > > > > > +
> > > > > > +	trace_xfs_btree_bload_level_geometry(cur, level, nr_this_level,
> > > > > > +			*avg_per_block, desired_npb, *blocks,
> > > > > > +			*blocks_with_extra);
> > > > > > +}
> > > > > > +
> > > > > > +/*
> > > > > > + * Ensure a slack value is appropriate for the btree.
> > > > > > + *
> > > > > > + * If the slack value is negative, set slack so that we fill the block to
> > > > > > + * halfway between minrecs and maxrecs.  Make sure the slack is never so large
> > > > > > + * that we can underflow minrecs.
> > > > > > + */
> > > > > > +static void
> > > > > > +xfs_btree_bload_ensure_slack(
> > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > +	int			*slack,
> > > > > > +	int			level)
> > > > > > +{
> > > > > > +	int			maxr;
> > > > > > +	int			minr;
> > > > > > +
> > > > > > +	/*
> > > > > > +	 * We only care about slack for btree blocks, so set the btree nlevels
> > > > > > +	 * to 3 so that level 0 is a leaf block and level 1 is a node block.
> > > > > > +	 * Avoid straying into inode roots, since we don't do slack there.
> > > > > > +	 */
> > > > > > +	cur->bc_nlevels = 3;
> > > > > > +	maxr = cur->bc_ops->get_maxrecs(cur, level);
> > > > > > +	minr = cur->bc_ops->get_minrecs(cur, level);
> > > > > > +
> > > > > > +	/*
> > > > > > +	 * If slack is negative, automatically set slack so that we load the
> > > > > > +	 * btree block approximately halfway between minrecs and maxrecs.
> > > > > > +	 * Generally, this will net us 75% loading.
> > > > > > +	 */
> > > > > > +	if (*slack < 0)
> > > > > > +		*slack = maxr - ((maxr + minr) >> 1);
> > > > > > +
> > > > > > +	*slack = min(*slack, maxr - minr);
> > > > > > +}
> > > > > > +
> > > > > > +/*
> > > > > > + * Prepare a btree cursor for a bulk load operation by computing the geometry
> > > > > > + * fields in @bbl.  Caller must ensure that the btree cursor is a staging
> > > > > > + * cursor.  This function can be called multiple times.
> > > > > > + */
> > > > > > +int
> > > > > > +xfs_btree_bload_compute_geometry(
> > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > +	struct xfs_btree_bload	*bbl,
> > > > > > +	uint64_t		nr_records)
> > > > > > +{
> > > > > > +	uint64_t		nr_blocks = 0;
> > > > > > +	uint64_t		nr_this_level;
> > > > > > +
> > > > > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > > > > +
> > > > > > +	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
> > > > > > +	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
> > > > > > +
> > > > > > +	bbl->nr_records = nr_this_level = nr_records;
> > > > > > +	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
> > > > > > +		uint64_t	level_blocks;
> > > > > > +		uint64_t	dontcare64;
> > > > > > +		unsigned int	level = cur->bc_nlevels - 1;
> > > > > > +		unsigned int	avg_per_block;
> > > > > > +
> > > > > > +		/*
> > > > > > +		 * If all the things we want to store at this level would fit
> > > > > > +		 * in a single root block, then we have our btree root and are
> > > > > > +		 * done.  Note that bmap btrees do not allow records in the
> > > > > > +		 * root.
> > > > > > +		 */
> > > > > > +		if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) || level != 0) {
> > > > > > +			xfs_btree_bload_level_geometry(cur, bbl, level,
> > > > > > +					nr_this_level, &avg_per_block,
> > > > > > +					&level_blocks, &dontcare64);
> > > > > > +			if (nr_this_level <= avg_per_block) {
> > > > > > +				nr_blocks++;
> > > > > > +				break;
> > > > > > +			}
> > > > > > +		}
> > > > > > +
> > > > > > +		/*
> > > > > > +		 * Otherwise, we have to store all the records for this level
> > > > > > +		 * in blocks and therefore need another level of btree to point
> > > > > > +		 * to those blocks.  Increase the number of levels and
> > > > > > +		 * recompute the number of records we can store at this level
> > > > > > +		 * because that can change depending on whether or not a level
> > > > > > +		 * is the root level.
> > > > > > +		 */
> > > > > > +		cur->bc_nlevels++;
> > > > > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > > > +				&avg_per_block, &level_blocks, &dontcare64);
> > > > > > +		nr_blocks += level_blocks;
> > > > > > +		nr_this_level = level_blocks;
> > > > > > +	}
> > > > > > +
> > > > > > +	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
> > > > > > +		return -EOVERFLOW;
> > > > > > +
> > > > > > +	bbl->btree_height = cur->bc_nlevels;
> > > > > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
> > > > > > +		bbl->nr_blocks = nr_blocks - 1;
> > > > > > +	else
> > > > > > +		bbl->nr_blocks = nr_blocks;
> > > > > > +	return 0;
> > > > > > +}
> > > > > > +
> > > > > > +/*
> > > > > > + * Bulk load a btree.
> > > > > > + *
> > > > > > + * Load @bbl->nr_records quantity of records into a btree using the supplied
> > > > > > + * empty and staging btree cursor @cur and a @bbl that has been filled out by
> > > > > > + * the xfs_btree_bload_compute_geometry function.
> > > > > > + *
> > > > > > + * The @bbl->get_data function must populate the cursor's bc_rec every time it
> > > > > > + * is called.  The @bbl->alloc_block function will be used to allocate new
> > > > > > + * btree blocks.  @priv is passed to both functions.
> > > > > > + *
> > > > > > + * Caller must ensure that @cur is a staging cursor.  Any existing btree rooted
> > > > > > + * in the fakeroot will be lost, so do not call this function twice.
> > > > > > + */
> > > > > > +int
> > > > > > +xfs_btree_bload(
> > > > > > +	struct xfs_btree_cur		*cur,
> > > > > > +	struct xfs_btree_bload		*bbl,
> > > > > > +	void				*priv)
> > > > > > +{
> > > > > > +	union xfs_btree_ptr		child_ptr;
> > > > > > +	union xfs_btree_ptr		ptr;
> > > > > > +	struct xfs_buf			*bp = NULL;
> > > > > > +	struct xfs_btree_block		*block = NULL;
> > > > > > +	uint64_t			nr_this_level = bbl->nr_records;
> > > > > > +	uint64_t			blocks;
> > > > > > +	uint64_t			i;
> > > > > > +	uint64_t			blocks_with_extra;
> > > > > > +	uint64_t			total_blocks = 0;
> > > > > > +	unsigned int			avg_per_block;
> > > > > > +	unsigned int			level = 0;
> > > > > > +	int				ret;
> > > > > > +
> > > > > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > > > > +
> > > > > > +	cur->bc_nlevels = bbl->btree_height;
> > > > > > +	xfs_btree_set_ptr_null(cur, &child_ptr);
> > > > > > +	xfs_btree_set_ptr_null(cur, &ptr);
> > > > > > +
> > > > > > +	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > > > +			&avg_per_block, &blocks, &blocks_with_extra);
> > > > > > +
> > > > > > +	/* Load each leaf block. */
> > > > > > +	for (i = 0; i < blocks; i++) {
> > > > > > +		unsigned int		nr_this_block = avg_per_block;
> > > > > > +
> > > > > > +		if (i < blocks_with_extra)
> > > > > > +			nr_this_block++;
> > > > > > +
> > > > > > +		ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > > > > +				nr_this_block, &ptr, &bp, &block, priv);
> > > > > > +		if (ret)
> > > > > > +			return ret;
> > > > > > +
> > > > > > +		trace_xfs_btree_bload_block(cur, level, i, blocks, &ptr,
> > > > > > +				nr_this_block);
> > > > > > +
> > > > > > +		ret = xfs_btree_bload_leaf(cur, nr_this_block, bbl->get_data,
> > > > > > +				block, priv);
> > > > > > +		if (ret)
> > > > > > +			goto out;
> > > > > > +
> > > > > > +		/* Record the leftmost pointer to start the next level. */
> > > > > > +		if (i == 0)
> > > > > > +			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
> > > > > > +	}
> > > > > > +	total_blocks += blocks;
> > > > > > +	xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > > > > +
> > > > > > +	/* Populate the internal btree nodes. */
> > > > > > +	for (level = 1; level < cur->bc_nlevels; level++) {
> > > > > > +		union xfs_btree_ptr	first_ptr;
> > > > > > +
> > > > > > +		nr_this_level = blocks;
> > > > > > +		block = NULL;
> > > > > > +		xfs_btree_set_ptr_null(cur, &ptr);
> > > > > > +
> > > > > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > > > +				&avg_per_block, &blocks, &blocks_with_extra);
> > > > > > +
> > > > > > +		/* Load each node block. */
> > > > > > +		for (i = 0; i < blocks; i++) {
> > > > > > +			unsigned int	nr_this_block = avg_per_block;
> > > > > > +
> > > > > > +			if (i < blocks_with_extra)
> > > > > > +				nr_this_block++;
> > > > > > +
> > > > > > +			ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > > > > +					nr_this_block, &ptr, &bp, &block,
> > > > > > +					priv);
> > > > > > +			if (ret)
> > > > > > +				return ret;
> > > > > > +
> > > > > > +			trace_xfs_btree_bload_block(cur, level, i, blocks,
> > > > > > +					&ptr, nr_this_block);
> > > > > > +
> > > > > > +			ret = xfs_btree_bload_node(cur, nr_this_block,
> > > > > > +					&child_ptr, block);
> > > > > > +			if (ret)
> > > > > > +				goto out;
> > > > > > +
> > > > > > +			/*
> > > > > > +			 * Record the leftmost pointer to start the next level.
> > > > > > +			 */
> > > > > > +			if (i == 0)
> > > > > > +				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
> > > > > > +		}
> > > > > > +		total_blocks += blocks;
> > > > > > +		xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > > > > +		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
> > > > > > +	}
> > > > > > +
> > > > > > +	/* Initialize the new root. */
> > > > > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
> > > > > > +		ASSERT(xfs_btree_ptr_is_null(cur, &ptr));
> > > > > > +		cur->bc_private.b.ifake->if_levels = cur->bc_nlevels;
> > > > > > +		cur->bc_private.b.ifake->if_blocks = total_blocks - 1;
> > > > > > +	} else {
> > > > > > +		cur->bc_private.a.afake->af_root = be32_to_cpu(ptr.s);
> > > > > > +		cur->bc_private.a.afake->af_levels = cur->bc_nlevels;
> > > > > > +		cur->bc_private.a.afake->af_blocks = total_blocks;
> > > > > > +	}
> > > > > > +out:
> > > > > > +	if (bp)
> > > > > > +		xfs_trans_brelse(cur->bc_tp, bp);
> > > > > > +	return ret;
> > > > > > +}
> > > > > > diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
> > > > > > index a17becb72ab8..5c6992a04ea2 100644
> > > > > > --- a/fs/xfs/libxfs/xfs_btree.h
> > > > > > +++ b/fs/xfs/libxfs/xfs_btree.h
> > > > > > @@ -582,4 +582,47 @@ void xfs_btree_stage_ifakeroot(struct xfs_btree_cur *cur,
> > > > > >  void xfs_btree_commit_ifakeroot(struct xfs_btree_cur *cur, int whichfork,
> > > > > >  		const struct xfs_btree_ops *ops);
> > > > > >  
> > > > > > +typedef int (*xfs_btree_bload_get_fn)(struct xfs_btree_cur *cur, void *priv);
> > > > > > +typedef int (*xfs_btree_bload_alloc_block_fn)(struct xfs_btree_cur *cur,
> > > > > > +		union xfs_btree_ptr *ptr, void *priv);
> > > > > > +typedef size_t (*xfs_btree_bload_iroot_size_fn)(struct xfs_btree_cur *cur,
> > > > > > +		unsigned int nr_this_level, void *priv);
> > > > > > +
> > > > > > +/* Bulk loading of staged btrees. */
> > > > > > +struct xfs_btree_bload {
> > > > > > +	/* Function to store a record in the cursor. */
> > > > > > +	xfs_btree_bload_get_fn		get_data;
> > > > > > +
> > > > > > +	/* Function to allocate a block for the btree. */
> > > > > > +	xfs_btree_bload_alloc_block_fn	alloc_block;
> > > > > > +
> > > > > > +	/* Function to compute the size of the in-core btree root block. */
> > > > > > +	xfs_btree_bload_iroot_size_fn	iroot_size;
> > > > > > +
> > > > > > +	/* Number of records the caller wants to store. */
> > > > > > +	uint64_t			nr_records;
> > > > > > +
> > > > > > +	/* Number of btree blocks needed to store those records. */
> > > > > > +	uint64_t			nr_blocks;
> > > > > > +
> > > > > > +	/*
> > > > > > +	 * Number of free records to leave in each leaf block.  If this (or
> > > > > > +	 * any of the slack values) are negative, this will be computed to
> > > > > > +	 * be halfway between maxrecs and minrecs.  This typically leaves the
> > > > > > +	 * block 75% full.
> > > > > > +	 */
> > > > > > +	int				leaf_slack;
> > > > > > +
> > > > > > +	/* Number of free keyptrs to leave in each node block. */
> > > > > > +	int				node_slack;
> > > > > > +
> > > > > > +	/* Computed btree height. */
> > > > > > +	unsigned int			btree_height;
> > > > > > +};
> > > > > > +
> > > > > > +int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
> > > > > > +		struct xfs_btree_bload *bbl, uint64_t nr_records);
> > > > > > +int xfs_btree_bload(struct xfs_btree_cur *cur, struct xfs_btree_bload *bbl,
> > > > > > +		void *priv);
> > > > > > +
> > > > > >  #endif	/* __XFS_BTREE_H__ */
> > > > > > diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
> > > > > > index bc85b89f88ca..9b5e58a92381 100644
> > > > > > --- a/fs/xfs/xfs_trace.c
> > > > > > +++ b/fs/xfs/xfs_trace.c
> > > > > > @@ -6,6 +6,7 @@
> > > > > >  #include "xfs.h"
> > > > > >  #include "xfs_fs.h"
> > > > > >  #include "xfs_shared.h"
> > > > > > +#include "xfs_bit.h"
> > > > > >  #include "xfs_format.h"
> > > > > >  #include "xfs_log_format.h"
> > > > > >  #include "xfs_trans_resv.h"
> > > > > > diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
> > > > > > index a78055521fcd..6d7ba64b7a0f 100644
> > > > > > --- a/fs/xfs/xfs_trace.h
> > > > > > +++ b/fs/xfs/xfs_trace.h
> > > > > > @@ -35,6 +35,7 @@ struct xfs_icreate_log;
> > > > > >  struct xfs_owner_info;
> > > > > >  struct xfs_trans_res;
> > > > > >  struct xfs_inobt_rec_incore;
> > > > > > +union xfs_btree_ptr;
> > > > > >  
> > > > > >  DECLARE_EVENT_CLASS(xfs_attr_list_class,
> > > > > >  	TP_PROTO(struct xfs_attr_list_context *ctx),
> > > > > > @@ -3670,6 +3671,90 @@ TRACE_EVENT(xfs_btree_commit_ifakeroot,
> > > > > >  		  __entry->blocks)
> > > > > >  )
> > > > > >  
> > > > > > +TRACE_EVENT(xfs_btree_bload_level_geometry,
> > > > > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > > > > +		 uint64_t nr_this_level, unsigned int nr_per_block,
> > > > > > +		 unsigned int desired_npb, uint64_t blocks,
> > > > > > +		 uint64_t blocks_with_extra),
> > > > > > +	TP_ARGS(cur, level, nr_this_level, nr_per_block, desired_npb, blocks,
> > > > > > +		blocks_with_extra),
> > > > > > +	TP_STRUCT__entry(
> > > > > > +		__field(dev_t, dev)
> > > > > > +		__field(xfs_btnum_t, btnum)
> > > > > > +		__field(unsigned int, level)
> > > > > > +		__field(unsigned int, nlevels)
> > > > > > +		__field(uint64_t, nr_this_level)
> > > > > > +		__field(unsigned int, nr_per_block)
> > > > > > +		__field(unsigned int, desired_npb)
> > > > > > +		__field(unsigned long long, blocks)
> > > > > > +		__field(unsigned long long, blocks_with_extra)
> > > > > > +	),
> > > > > > +	TP_fast_assign(
> > > > > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > > > > +		__entry->btnum = cur->bc_btnum;
> > > > > > +		__entry->level = level;
> > > > > > +		__entry->nlevels = cur->bc_nlevels;
> > > > > > +		__entry->nr_this_level = nr_this_level;
> > > > > > +		__entry->nr_per_block = nr_per_block;
> > > > > > +		__entry->desired_npb = desired_npb;
> > > > > > +		__entry->blocks = blocks;
> > > > > > +		__entry->blocks_with_extra = blocks_with_extra;
> > > > > > +	),
> > > > > > +	TP_printk("dev %d:%d btree %s level %u/%u nr_this_level %llu nr_per_block %u desired_npb %u blocks %llu blocks_with_extra %llu",
> > > > > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > > > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > > > > +		  __entry->level,
> > > > > > +		  __entry->nlevels,
> > > > > > +		  __entry->nr_this_level,
> > > > > > +		  __entry->nr_per_block,
> > > > > > +		  __entry->desired_npb,
> > > > > > +		  __entry->blocks,
> > > > > > +		  __entry->blocks_with_extra)
> > > > > > +)
> > > > > > +
> > > > > > +TRACE_EVENT(xfs_btree_bload_block,
> > > > > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > > > > +		 uint64_t block_idx, uint64_t nr_blocks,
> > > > > > +		 union xfs_btree_ptr *ptr, unsigned int nr_records),
> > > > > > +	TP_ARGS(cur, level, block_idx, nr_blocks, ptr, nr_records),
> > > > > > +	TP_STRUCT__entry(
> > > > > > +		__field(dev_t, dev)
> > > > > > +		__field(xfs_btnum_t, btnum)
> > > > > > +		__field(unsigned int, level)
> > > > > > +		__field(unsigned long long, block_idx)
> > > > > > +		__field(unsigned long long, nr_blocks)
> > > > > > +		__field(xfs_agnumber_t, agno)
> > > > > > +		__field(xfs_agblock_t, agbno)
> > > > > > +		__field(unsigned int, nr_records)
> > > > > > +	),
> > > > > > +	TP_fast_assign(
> > > > > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > > > > +		__entry->btnum = cur->bc_btnum;
> > > > > > +		__entry->level = level;
> > > > > > +		__entry->block_idx = block_idx;
> > > > > > +		__entry->nr_blocks = nr_blocks;
> > > > > > +		if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
> > > > > > +			xfs_fsblock_t	fsb = be64_to_cpu(ptr->l);
> > > > > > +
> > > > > > +			__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsb);
> > > > > > +			__entry->agbno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsb);
> > > > > > +		} else {
> > > > > > +			__entry->agno = cur->bc_private.a.agno;
> > > > > > +			__entry->agbno = be32_to_cpu(ptr->s);
> > > > > > +		}
> > > > > > +		__entry->nr_records = nr_records;
> > > > > > +	),
> > > > > > +	TP_printk("dev %d:%d btree %s level %u block %llu/%llu fsb (%u/%u) recs %u",
> > > > > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > > > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > > > > +		  __entry->level,
> > > > > > +		  __entry->block_idx,
> > > > > > +		  __entry->nr_blocks,
> > > > > > +		  __entry->agno,
> > > > > > +		  __entry->agbno,
> > > > > > +		  __entry->nr_records)
> > > > > > +)
> > > > > > +
> > > > > >  #endif /* _TRACE_XFS_H */
> > > > > >  
> > > > > >  #undef TRACE_INCLUDE_PATH
> > > > > >

Brian Foster Oct. 18, 2019, 2:38 p.m. UTC | #7

On Thu, Oct 17, 2019 at 12:06:06PM -0700, Darrick J. Wong wrote:
> On Thu, Oct 17, 2019 at 05:32:59AM -0400, Brian Foster wrote:
> > On Wed, Oct 16, 2019 at 05:40:18PM -0700, Darrick J. Wong wrote:
> > > On Wed, Oct 16, 2019 at 05:07:31PM -0400, Brian Foster wrote:
> > > > On Wed, Oct 16, 2019 at 11:15:02AM -0700, Darrick J. Wong wrote:
> > > > > On Wed, Oct 16, 2019 at 11:26:48AM -0400, Brian Foster wrote:
> > > > > > On Wed, Oct 09, 2019 at 09:48:18AM -0700, Darrick J. Wong wrote:
> > > > > > > From: Darrick J. Wong <[e-mail]>
> > > > > > > 
> > > > > > > Add a new btree function that enables us to bulk load a btree cursor.
> > > > > > > This will be used by the upcoming online repair patches to generate new
> > > > > > > btrees.
> > > > > > > 
> > > > > > > Signed-off-by: Darrick J. Wong <[e-mail]>
> > > > > > > ---
> > > > > > >  fs/xfs/libxfs/xfs_btree.c |  566 +++++++++++++++++++++++++++++++++++++++++++++
> > > > > > >  fs/xfs/libxfs/xfs_btree.h |   43 +++
> > > > > > >  fs/xfs/xfs_trace.c        |    1 
> > > > > > >  fs/xfs/xfs_trace.h        |   85 +++++++
> > > > > > >  4 files changed, 694 insertions(+), 1 deletion(-)
> > > > > > > 
> > > > > > > 
> > > > > > > diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
> > > > > > > index 4b06d5d86834..17b0fdb87729 100644
> > > > > > > --- a/fs/xfs/libxfs/xfs_btree.c
> > > > > > > +++ b/fs/xfs/libxfs/xfs_btree.c
> > > > > > ...
> > > > > > > @@ -5104,3 +5104,567 @@ xfs_btree_commit_ifakeroot(
> > > > > > >  	cur->bc_ops = ops;
> > > > > > >  	cur->bc_flags &= ~XFS_BTREE_STAGING;
> > > > > > >  }
> > > > > > > +
> > > > > > > +/*
> > > > > > > + * Bulk Loading of Staged Btrees
> > > > > > > + * =============================
> > > > > > > + *
> > > > > > > + * This interface is used with a staged btree cursor to create a totally new
> > > > > > > + * btree with a large number of records (i.e. more than what would fit in a
> > > > > > > + * single block).  When the creation is complete, the new root can be linked
> > > > > > > + * atomically into the filesystem by committing the staged cursor.
> > > > > > > + *
> > > > > 
> > > > > [paraphrasing a conversation we had on irc]
> > > > > 
> > > > > > Thanks for the documentation. So what is the purpose behind the whole
> > > > > > bulk loading thing as opposed to something like faking up an AG
> > > > > > structure (i.e. AGF) somewhere and using the existing cursor mechanisms
> > > > > > (or something closer to it) to copy records from one place to another?
> > > > > > Is it purely a performance/efficiency tradeoff? Bulk block allocation
> > > > > > issues? Transactional/atomicity issues? All (or none :P) of the above?
> > > > > 
> > > > > Prior to the v20, the online repair series created a new btree root,
> > > > > committed that into wherever the root lived, and inserted records one by
> > > > > one into the btree.  There were quite a few drawbacks to this method:
> > > > > 
> > > > > 1. Inserting records one at a time can involve walking up the tree to
> > > > > update node block pointers, which isn't terribly efficient if we're
> > > > > likely going to rewrite the pointers (and relogging nodes) several more
> > > > > times.
> > > > > 
> > > > > 2. Inserting records one at a time tends to leave a lot of half-empty
> > > > > btree blocks because when one block fills up we split it and push half
> > > > > the records to the new block.  It would be nice not to explode the size
> > > > > of the btrees, and it would be particularly useful if we could control
> > > > > the load factor of the new btree precisely.
> > > > > 
> > > > 
> > > > Interesting... this is a trait the traditional btree update paths share
> > > > though, right?
> > > 
> > > Right.  It's similar to the behavior Dave was seeing a couple of weeks
> > > ago with Zorro's stress testing of the incore extent cache.
> > > 
> > > > > 3. The rebuild wasn't atomic, since we were replacing the root prior to
> > > > > the insert loop.  If we crashed midway through a rebuild we'd end up
> > > > > with a garbage btree and no indication that it was incorrect.  That's
> > > > > how the fakeroot code got started.
> > > > > 
> > > > 
> > > > Indeed, though this seems more related to the anchoring (i.e. fake root)
> > > > approach than bulk vs. iterative construction.
> > > 
> > > Correct.
> > > 
> > > > > 4. In a previous version of the repair series I tried to batch as many
> > > > > insert operations into a single transaction as possible, but my
> > > > > transaction reservation fullness estimation function didn't work
> > > > > reliably (particularly when things got really fragmented), so I backed
> > > > > off to rolling after /every/ insertion.  That works well enough, but at
> > > > > a cost of a lot of transaction rolling, which means that repairs plod
> > > > > along very slowly.
> > > > > 
> > > > > 5. Performing an insert loop means that the btree blocks are allocated
> > > > > one at a time as the btree expands.  This is suboptimal since we can
> > > > > calculate the exact size of the new btree prior to building it, which
> > > > > gives us the opportunity to recreate the index in a set of contiguous
> > > > > blocks instead of scattering them.
> > > > > 
> > > > 
> > > > Yep, FWIW it sounds like most of these tradeoffs are around
> > > > performance/efficiency. 
> > > 
> > > <nod>
> > > 
> > > > > 6. If we crash midway through a rebuild, XFS neither cleaned up the mess
> > > > > nor informed the administrator that it was necessary to re-run xfs_scrub
> > > > > or xfs_repair to clean up the lost blocks.  Obviously, automatic cleanup
> > > > > is a far better solution.
> > > > > 
> > > > 
> > > > Similar to above, I think this kind of depends more on how/where to
> > > > anchor an in-progress tree as opposed to what level records are copied
> > > > at.
> > > 
> > > <nod> The six points are indeed the overall list of complaints about the
> > > v19 code. :)
> > > 
> > > > > The first thing I decided to solve was the lack of atomicity.
> > > > > 
> > > > > For AG-rooted btrees, I thought about creating a fake xfs_buf for an AG
> > > > > header buffer and extracting the root/level values after construction
> > > > > completes.  That's possible, but it's risky because the fake buffer
> > > > > could get logged and if the sector number matches the actual header
> > > > > then it introduces buffer cache aliasing issues.
> > > > > 
> > > > > For inode-rooted btrees, one could create a fake xfs_inode with the same
> > > > > i_ino as the target.  That presents the same aliasing issues as the fake
> > > > > xfs_buf above.  A different strategy would be to allocate an unlinked
> > > > > inode and then use the bmbt owner change (a.k.a. extent swap) to move
> > > > > the mappings over.  That would work, though it has two large drawbacks:
> > > > > (a) a lot of additional complexity around allocating and freeing the
> > > > > temporary inode; and (b) future inode-rooted btrees such as the realtime
> > > > > rmap btree would also have to implement an owner-change operation.
> > > > > 
> > > > 
> > > > I was wondering more along the lines of having an actual anchor
> > > > somewhere. E.g., think of it as a temporary/inaccessible location of a
> > > > legitimate on-disk structure as opposed to a fake object in memory
> > > > somewhere. A hidden/internal repair inode or some such, perhaps. I'm
> > > > sure there's new code/complexity that would come around with that, but I
> > > > think that's going to be unavoidable to some degree for an online repair
> > > > mechanism. ;)
> > > 
> > > <nod> So far I /think/ I've managed to keep to an absolute minimum the
> > > amount of metadata that gets written to disk prior to the commit.  I
> > > haven't reread the series with an eye for how v20 is going to come up
> > > short though. :)
> > > 
> > 
> > This metadata has to be written out one way or another though. IOW, it
> > seems the advantage to holding off is mostly for the case where we crash
> > or something mid reconstruction (which should be the uncommon case :P).
> 
> Right.
> 
> > I could see how there might be some benefits to using contiguous blocks
> > in a btree and batching them out as such, but conversely I'm also
> > curious about how that might behave with certain scenarios like really
> > large btrees that take a while to reconstruct (e.g. due to read
> > contention or slow storage) or where contiguous blocks aren't available,
> > etc.
> 
> You mean like a debug knob or something to force the allocation to
> scatter blocks everywhere?  I do some regular testing of large(r) btree
> reconstruction with a metadump of /home that's been mdrestored to a
> thinp device.
> 

I don't think that is really necessary. I just mean that it's easy to
create badly fragmented free space conditions and we have to accommodate
that case regardless. Hence, there's a reasonable cost/benefit question
between submitting all buffer I/O at once, submitting each buffer as
it's populated, or anything in between when we consider the spectrum of
possible conditions from ideal (contiguous btree blocks) to worst case
(one block per allocation).

> (Or I guess building an xfstests that sets up a slow dm device...?)
> 

Yep.. For example, it's fairly clear how having a nice set of contiguous
blocks facilitates writing out the reconstructed btree all at once. But
what if we create a large filesystem, fragment free space, possibly slow
down the device as noted, then reconstruct a worst case (fragmented)
btree with a competing userspace workload. How does a "write everything
at once" approach to loading the btree fare under those kind of worst
case conditions as compared to the ideal approach (or as compared to
something that might submit btree blocks for I/O as they are populated
or in smaller batches)?

> > > I may very well have to revisit the hidden/internal repair inode concept
> > > whenever I start working on rebuilding directories and xattrs since I
> > > can't see any other way of atomically rebuilding those.  But that's
> > > very very far out still.
> > > 
> > 
> > Ok.
> > 
> > > > Note that this is all just handwaving on my part and still without full
> > > > context as to how things are currently anchored, made atomic, etc. I'm
> > > > primarily trying to understand the design reasoning based on the high
> > > > level description.
> > > 
> > > <nod>
> > > 
> > > > > To fix (3), I thought it wise to have explicit fakeroot structures to
> > > > > maintain a clean separation between what we're building and the rest of
> > > > > the filesystem.  This also means that there's nothing on disk to clean
> > > > > up if we fail at any point before we're ready to commit the new btree.
> > > > > 
> > > > 
> > > > Hmm.. so this approach facilites a tree reconstruction in a single open
> > > > transaction? If so, I suppose I could see some functional advantages to
> > > > that.
> > > 
> > > Correct.
> > > 
> > > > > Then Dave (I think?) suggested that I  use EFIs strategically to
> > > > > schedule freeing of the new btree blocks (the root commit transaction
> > > > > would log EFDs to cancel them) and to schedule freeing of the old
> > > > > blocks.  That solves (6), though the EFI wrangling doesn't happen for
> > > > > another couple of series after this one.
> > > > > 
> > > > 
> > > > Hm, Ok... so new btree block allocation(s?) in the same transaction as
> > > > an EFI, to be processed on recovery if we crash, otherwise cancelled
> > > > with an EFD on construction completion..?
> > > 
> > > Correct.  In the end, the transaction sequence looks like:
> > > 
> > > T[1]: Allocate an extent, log metadata updates to reflect that, log EFI
> > > for the extent.
> > > 
> > > <repeat until we've allocated as many blocks as we need>
> > > 
> > 
> > If I follow correctly, I take it "repeat" here means we're rolling the
> > transaction to reallocate, and relog all previous EFIs along the way,
> > until we've acquired sufficient blocks to start populating the tree.
> 
> Right, though I don't explicitly relog the previous EFIs.
> 

Ok.

> > > T[N]: Attach ordered buffers for the new btree's blocks.  Log the root
> > > change.  Log EFDs for all the EFIs logged in T[1..N-1].  Log EFIs for
> > > all the old btree blocks that we could find.
> > > 
> > 
> > So here we're on/after the last allocation transaction, have a bunch of
> > pending EFIs for recovery purposes, and start populating the btree
> > blocks...
> > 
> > AFAICT the log tail is pinned by these EFIs during population, which is
> > of non-deterministic time for aforementioned reasons. If so, this
> > _sounds_ like it risks a similar log deadlock vector to the one we
> > currently have for quotaoff, though I could easily be missing something
> > from the high level description.
> 
> Yes, the EFIs pin the log and put us at risk for a quotaoff style log
> deadlock.  This was particularly risky with the insert_rec loop since
> repair itself was contributing to log traffic.  I've wondered what would
> happen if we could relog intent items the same way we do with inode
> cores to avoid pinning the tail?
> 

You mean like an automatic intent relogging mechanism built into the
log? That sounds interesting, but I'd have to think about that some
more. Something like that probably also solves the longstanding quotaoff
thing too, FWIW.

> I guess it's /not/ that similar since the log itself could relog the
> pinned intent items since they're not supposed to change once they've
> been logged the first time.
> 

Hmm.. the most straightforward approach to relogging is to re-add the
associated item to the next transaction because that accounts for log
reservation in the CIL context ticket, etc. The problem with such
explicit relogging in cases like quotaoff is that (IIRC) we have
codepaths through various twisty paths such as generic structure
iteraters, we could release an inode, go the vfs and back into XFS to
allocate a transaction, etc. where it's not really practical to pass
around the log item for relogging. Deferred operations facilitates this
kind of automatic relogging because transaction rolling is an inherent
requirement.

I think the ideal solution for that problem would be to _somehow_ be
able to flag an intent when it is initially committed to cause it to be
reinserted to the CIL on every checkpoint. That means it still goes into
the AIL as normal, but is essentially immediately and automatically
relogged in the next CIL checkpoint. The immediate challenge with that
is dealing with the log reservation required to relog said item in the
next checkpoint, because all of the CIL checkpoint reservation comes
from transaction reservations. We also need to find a way to allocate a
new log vector for the item, since that happens outside of the CIL lock
to avoid deadlocks IIRC. Another more subtle issue is the question of
whether we want/need to still have the ability to cover the log with an
outstanding "automatic relog" intent (or conversely, not be constantly
spinning the log covering wq on a single item), but that may or may not
be a problem.

Hmmmmmm.. I'd have to think about this some more to see if there's a
clever way to steal more reservation such that relogging is accurate and
reliable. Since these intents should be mostly pretty small, I'm
wondering if we could do something like maintain a list of pending items
in the CIL and steal from the next transaction commit that can support a
relog. So when the target item is first committed it goes through the
normal log item lifecycle, except once a checkpoint occurs the item goes
in the AIL and a private CIL list awaiting relog. The first transaction
that commits after that point with enough extra reservation to support a
relog of item(s) on the pending list, steals as many off the list as it
can and commits them as part of the current transaction. Then the items
are back on the CIL, off the pending list, relog in the next checkpoint
and the process repeats until they are ultimately completed (with an
intent done). Thoughts on that? Is that along the lines of what you were
thinking..?

If so, perhaps I'll play around with this in the context of quotaoff
since I never really came up with a workable solution last time I tried.
That would at least potentially fix an existing issue and also tell us
whether this is something we could use for repair design going forward.

> > Also, what happens if we complete the population, roll this transaction,
> > unpin the ordered buffers and crash before the entire set of ordered
> > buffers is written back? We currently use ordered buffers with
> > operations described by logical intents. For example, an icreate item
> > goes into the log with an ordered buffer such that if we crash before
> > buffer writeback completes and allows the tail to move, recovery knows
> > how to turn the active icreate log item back into a valid inode chunk.
> > Is there something similar going on here?
> 
> Ooooh, yes.  Paraphrasing (and expanding upon) what discussed on IRC
> earlier, I forgot that ordered buffers are written out by the AIL, which
> means that xfs_btree_bload needs to force the new blocks to be written
> to disk before logging and committing the new root.
> 

Yeah, that probably makes more sense.

> Ok, so I guess we need xfs_btree_bload to delwri_queue each new block's
> buffer into a list and then delwri_submit the list before returning.  If
> we hit an error we'll cancel the transaction, if we succeed then we move
> on to commit the staging cursor, etc.
> 
> Hmm, thinking about it further, if the delwri_submit fails we could
> commit the transaction and return from repair without committing the
> root.  The EFIs will trigger, which frees the blocks we allocated, and
> the fs ends in the same place it was before the repair started.
> 
> Though that's an optimization; for now we'll just cancel the
> transaction and bail out.
> 

I view that as more of a simplification than an optimization. That's a
lot of allocating+freeing work to be done for a repair operation that
ultimately didn't do anything. ;) That said, I think it's reasonable to
focus on simplicity and functional correctness over the most optimal
error path. Things like failure handling can be improved down the road,
so long as we do something correct enough in the meantime so as to not
damage the fs. :)

> > > <roll transaction to write the ordered buffers and commit>
> > > 
> > > T[N+1]: Free an extent to finish the first EFI logged in the previous step.
> > > 
> > > <repeat until we've processed everything from the second wave of EFIs>
> > > 
> > 
> > And I take repeat here means essentially rolling and completing the
> > "real" EFIs for the old btree we've swapped out.
> 
> Yes.
> 
> > FWIW, it seems to me that this altogether might be an opportunity for a
> > custom logical intent to cover reconstruction sequences. 
> 
> How do you mean?  A new "btree reconstruction intent" log item?  Or
> maybe more generally, a "repair intent" log item?  Hmm, that would force
> a new log_incompat flag, but I'll think about that.  So far my thinking
> has been that a failed online repair should be followed up with an
> xfs_repair run, just in case the repair failed due to software bug.  But
> maybe recovery of the new intent type would log a message and return
> error to force xfs_repair, and when we're more confident we can change
> it to call into xfs_scrub_metadata().
> 

Well I hadn't thought about it that deeply, but what I was thinking in
this particular case was more of a btree reconstruction intent that
tracks the fact that a reconstruction is in progress. That facilitates a
physical anchor for allocated btree blocks in that the intent tells
recovery to clean things up. Another purpose behind that is we can just
use regular transactions to allocate btree blocks (no EFI games), log
btree buffer content (no pending in-core buffer lists), free the old
btree on completion, free the newly allocated blocks in the event of
failure, etc. IOW, we're back to using existing transactions for
incremental changes to maintain filesystem consistency and design new
logical transactions to track higher level repair operations.

In reality, one logical btree reconstruction intent might not be enough.
We may need multiple such intents to track reconstruction, the btree
commit, clean up of old blocks, etc. etc. I don't really have enough
context to reason about that, TBH. OTOH, I think that kind of approach
would essentially require breaking down the repair operations into
logical/generic components, which from a design standpoint IMO makes the
design a little easier to reason about. Anyways, this is all just
thoughts and handwaving on my part..

Brian

> > > Call xfs_trans_commit and we're done.
> > > 
> > > > > He also suggested using ordered buffers to write out the new btree
> > > > > blocks along with whatever logging was necessary to commit the new
> > > > > btree.  It then occurred to me that xfs_repair open-codes the process of
> > > > > calculating the geometry of a new btree, allocating all the blocks at
> > > > > once, and writing out full btree blocks.  Somewhat annoyingly, it
> > > > > features nearly the same (open-)code for all four AG btree types, which
> > > > > is less maintainable than it could be.
> > > > > 
> > > > > I read through all four versions and used it to write the generic btree
> > > > > bulk loading code.  For scrub I hooked that up to the "staged btree with
> > > > > a fake root" stuff I'd already written, which solves (1), (2), (4), and
> > > > > (5).
> > > > > 
> > > > > For xfsprogs[1], I deleted a few thousand lines of code from xfs_repair.
> > > > > True, we don't reuse existing common code, but we at least get to share
> > > > > new common btree code.
> > > > > 
> > > > 
> > > > Yeah, the xfsprogs work certainly makes sense. Part of the reason I ask
> > > > about this is the tradeoff of having multiple avenues to construct a
> > > > tree in the kernel codebase.
> > > 
> > > <nod>
> > > 
> > > > > > This is my first pass through this so I'm mostly looking at big picture
> > > > > > until I get to a point to see how these bits are used. The mechanism
> > > > > > itself seems reasonable in principle, but the reason I ask is it also
> > > > > > seems like there's inherent value in using more of same infrastructure
> > > > > > to reconstruct a tree that we use to create one in the first place. We
> > > > > > also already have primitives for things like fork swapping via the
> > > > > > extent swap mechanism, etc.
> > > > > 
> > > > > "bfoster: I guess it would be nice to see that kind of make it work ->
> > > > > make it fast evolution in tree"
> > > > > 
> > > > > For a while I did maintain the introduction of the bulk loading code as
> > > > > separate patches against the v19 repair code, but unfortunately I
> > > > > smushed them down before sending v20 to reduce the patch count, and
> > > > > because I didn't want to argue with everyone over the semi-working code
> > > > > that would then be replaced in the very next patch.
> > > > > 
> > > > 
> > > > That's not quite what I meant... The approach you've taken makes sense
> > > > to me for an implementation presented in a single series. I was more
> > > > thinking that at the point where it was determined the implementation
> > > > was going to change so drastically, after so many iterations it might
> > > > have been useful to see the v19 approach merged in an experimental form
> > > 
> > > I would have liked to see the online repair stuff merged in experimental
> > > form too so I can reduce the size of my patch queue, but oh well. :)
> > > 
> > 
> > Oh well, indeed. :P
> 
> Well, admittedly, it's /still/ a lot of half-baked code. :D
> 
> > > The silver lining to these lengthy reworks and slow review is that I can
> > > come back and do a fresh self-review after a month and straighten out
> > > the ugly parts as time goes by.  Unfortunately, that doesn't leave much
> > > of a paper trail or obvious evidence of development history.
> > > 
> > 
> > Exactly.
> > 
> > > Eight months ago it occurred to me that perhaps there is some value in
> > > retaining *some* periodic development history of this, so I've been
> > > adding dated tags to my integration repo[1] based on my development
> > > branch names, so I guess you could actually clone the git repo and git
> > > diff from one tag to another.  In general I'll generate a new pile of
> > > tags just before patchbombing.
> > > 
> > 
> > More importantly, I think we should try to follow that approach if
> > possible from here on, even if there are open/unresolved issues with the
> > currently proposed approach. Online repair overall is complex enough
> > that it would probably be impossible to get a fully functional online
> > repair done in a single series (and we're clearly not taking that
> > approach anyways), making sure that the various common underpinnings
> > work for all of the various types of structures, etc. Just my .02
> > though..
> 
> <nod>  I'll keep posting the tags and whatnot to my private repo and
> leave them there.  Anyone else pushing large patchsets could do
> likewise.
> 
> --D
> 
> > Brian
> > 
> > > (Granted 'repair-part-one' has been split into smaller parts now...)
> > > 
> > > > and then reworked upstream from there. Now that the new approach is
> > > > implemented, I agree it's probably not worth reinserting the old
> > > > approach at this point just to switch it out.
> > > 
> > > <nod>
> > > 
> > > > Thanks for the breakdown...
> > > 
> > > No problem, thanks for reading! :)
> > > 
> > > --D
> > > 
> > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux.git
> > > 
> > > > Brian
> > > > 
> > > > > I could split them back out, though at a cost of having to reintroduce a
> > > > > lot of hairy code in the bnobt/cntbt rebuild function to seed the free
> > > > > new space btree root in order to make sure that the btree block
> > > > > allocation code works properly, along with auditing the allocation paths
> > > > > to make sure they don't use the old AGF or encounter other subtleties.
> > > > > 
> > > > > It'd be a lot of work considering that the v20 reconstruction code is
> > > > > /much/ simpler than v19's was.  I also restructured the repair functions
> > > > > to allocate one large context structure at the beginning instead of the
> > > > > piecemeal way it was done onstack in v19 because stack usage was growing
> > > > > close to 1k in some cases.
> > > > > 
> > > > > --D
> > > > > 
> > > > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/log/?h=repair-bulk-load
> > > > > 
> > > > > > 
> > > > > > Brian
> > > > > > 
> > > > > > > + * The first step for the caller is to construct a fake btree root structure
> > > > > > > + * and a staged btree cursor.  A staging cursor contains all the geometry
> > > > > > > + * information for the btree type but will fail all operations that could have
> > > > > > > + * side effects in the filesystem (e.g. btree shape changes).  Regular
> > > > > > > + * operations will not work unless the staging cursor is committed and becomes
> > > > > > > + * a regular cursor.
> > > > > > > + *
> > > > > > > + * For a btree rooted in an AG header, use an xbtree_afakeroot structure.
> > > > > > > + * This should be initialized to zero.  For a btree rooted in an inode fork,
> > > > > > > + * use an xbtree_ifakeroot structure.  @if_fork_size field should be set to
> > > > > > > + * the number of bytes available to the fork in the inode; @if_fork should
> > > > > > > + * point to a freshly allocated xfs_inode_fork; and @if_format should be set
> > > > > > > + * to the appropriate fork type (e.g. XFS_DINODE_FMT_BTREE).
> > > > > > > + *
> > > > > > > + * The next step for the caller is to initialize a struct xfs_btree_bload
> > > > > > > + * context.  The @nr_records field is the number of records that are to be
> > > > > > > + * loaded into the btree.  The @leaf_slack and @node_slack fields are the
> > > > > > > + * number of records (or key/ptr) slots to leave empty in new btree blocks.
> > > > > > > + * If a caller sets a slack value to -1, the slack value will be computed to
> > > > > > > + * fill the block halfway between minrecs and maxrecs items per block.
> > > > > > > + *
> > > > > > > + * The number of items placed in each btree block is computed via the following
> > > > > > > + * algorithm: For leaf levels, the number of items for the level is nr_records.
> > > > > > > + * For node levels, the number of items for the level is the number of blocks
> > > > > > > + * in the next lower level of the tree.  For each level, the desired number of
> > > > > > > + * items per block is defined as:
> > > > > > > + *
> > > > > > > + * desired = max(minrecs, maxrecs - slack factor)
> > > > > > > + *
> > > > > > > + * The number of blocks for the level is defined to be:
> > > > > > > + *
> > > > > > > + * blocks = nr_items / desired
> > > > > > > + *
> > > > > > > + * Note this is rounded down so that the npb calculation below will never fall
> > > > > > > + * below minrecs.  The number of items that will actually be loaded into each
> > > > > > > + * btree block is defined as:
> > > > > > > + *
> > > > > > > + * npb =  nr_items / blocks
> > > > > > > + *
> > > > > > > + * Some of the leftmost blocks in the level will contain one extra record as
> > > > > > > + * needed to handle uneven division.  If the number of records in any block
> > > > > > > + * would exceed maxrecs for that level, blocks is incremented and npb is
> > > > > > > + * recalculated.
> > > > > > > + *
> > > > > > > + * In other words, we compute the number of blocks needed to satisfy a given
> > > > > > > + * loading level, then spread the items as evenly as possible.
> > > > > > > + *
> > > > > > > + * To complete this step, call xfs_btree_bload_compute_geometry, which uses
> > > > > > > + * those settings to compute the height of the btree and the number of blocks
> > > > > > > + * that will be needed to construct the btree.  These values are stored in the
> > > > > > > + * @btree_height and @nr_blocks fields.
> > > > > > > + *
> > > > > > > + * At this point, the caller must allocate @nr_blocks blocks and save them for
> > > > > > > + * later.  If space is to be allocated transactionally, the staging cursor
> > > > > > > + * must be deleted before and recreated after, which is why computing the
> > > > > > > + * geometry is a separate step.
> > > > > > > + *
> > > > > > > + * The fourth step in the bulk loading process is to set the function pointers
> > > > > > > + * in the bload context structure.  @get_data will be called for each record
> > > > > > > + * that will be loaded into the btree; it should set the cursor's bc_rec
> > > > > > > + * field, which will be converted to on-disk format and copied into the
> > > > > > > + * appropriate record slot.  @alloc_block should supply one of the blocks
> > > > > > > + * allocated in the previous step.  For btrees which are rooted in an inode
> > > > > > > + * fork, @iroot_size is called to compute the size of the incore btree root
> > > > > > > + * block.  Call xfs_btree_bload to start constructing the btree.
> > > > > > > + *
> > > > > > > + * The final step is to commit the staging cursor, which logs the new btree
> > > > > > > + * root and turns the btree into a regular btree cursor, and free the fake
> > > > > > > + * roots.
> > > > > > > + */
> > > > > > > +
> > > > > > > +/*
> > > > > > > + * Put a btree block that we're loading onto the ordered list and release it.
> > > > > > > + * The btree blocks will be written when the final transaction swapping the
> > > > > > > + * btree roots is committed.
> > > > > > > + */
> > > > > > > +static void
> > > > > > > +xfs_btree_bload_drop_buf(
> > > > > > > +	struct xfs_trans	*tp,
> > > > > > > +	struct xfs_buf		**bpp)
> > > > > > > +{
> > > > > > > +	if (*bpp == NULL)
> > > > > > > +		return;
> > > > > > > +
> > > > > > > +	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_BTREE_BUF);
> > > > > > > +	xfs_trans_ordered_buf(tp, *bpp);
> > > > > > > +	xfs_trans_brelse(tp, *bpp);
> > > > > > > +	*bpp = NULL;
> > > > > > > +}
> > > > > > > +
> > > > > > > +/* Allocate and initialize one btree block for bulk loading. */
> > > > > > > +STATIC int
> > > > > > > +xfs_btree_bload_prep_block(
> > > > > > > +	struct xfs_btree_cur		*cur,
> > > > > > > +	struct xfs_btree_bload		*bbl,
> > > > > > > +	unsigned int			level,
> > > > > > > +	unsigned int			nr_this_block,
> > > > > > > +	union xfs_btree_ptr		*ptrp,
> > > > > > > +	struct xfs_buf			**bpp,
> > > > > > > +	struct xfs_btree_block		**blockp,
> > > > > > > +	void				*priv)
> > > > > > > +{
> > > > > > > +	union xfs_btree_ptr		new_ptr;
> > > > > > > +	struct xfs_buf			*new_bp;
> > > > > > > +	struct xfs_btree_block		*new_block;
> > > > > > > +	int				ret;
> > > > > > > +
> > > > > > > +	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
> > > > > > > +	    level == cur->bc_nlevels - 1) {
> > > > > > > +		struct xfs_ifork	*ifp = cur->bc_private.b.ifake->if_fork;
> > > > > > > +		size_t			new_size;
> > > > > > > +
> > > > > > > +		/* Allocate a new incore btree root block. */
> > > > > > > +		new_size = bbl->iroot_size(cur, nr_this_block, priv);
> > > > > > > +		ifp->if_broot = kmem_zalloc(new_size, 0);
> > > > > > > +		ifp->if_broot_bytes = (int)new_size;
> > > > > > > +		ifp->if_flags |= XFS_IFBROOT;
> > > > > > > +
> > > > > > > +		/* Initialize it and send it out. */
> > > > > > > +		xfs_btree_init_block_int(cur->bc_mp, ifp->if_broot,
> > > > > > > +				XFS_BUF_DADDR_NULL, cur->bc_btnum, level,
> > > > > > > +				nr_this_block, cur->bc_private.b.ip->i_ino,
> > > > > > > +				cur->bc_flags);
> > > > > > > +
> > > > > > > +		*bpp = NULL;
> > > > > > > +		*blockp = ifp->if_broot;
> > > > > > > +		xfs_btree_set_ptr_null(cur, ptrp);
> > > > > > > +		return 0;
> > > > > > > +	}
> > > > > > > +
> > > > > > > +	/* Allocate a new leaf block. */
> > > > > > > +	ret = bbl->alloc_block(cur, &new_ptr, priv);
> > > > > > > +	if (ret)
> > > > > > > +		return ret;
> > > > > > > +
> > > > > > > +	ASSERT(!xfs_btree_ptr_is_null(cur, &new_ptr));
> > > > > > > +
> > > > > > > +	ret = xfs_btree_get_buf_block(cur, &new_ptr, &new_block, &new_bp);
> > > > > > > +	if (ret)
> > > > > > > +		return ret;
> > > > > > > +
> > > > > > > +	/* Initialize the btree block. */
> > > > > > > +	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
> > > > > > > +	if (*blockp)
> > > > > > > +		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
> > > > > > > +	xfs_btree_set_sibling(cur, new_block, ptrp, XFS_BB_LEFTSIB);
> > > > > > > +	xfs_btree_set_numrecs(new_block, nr_this_block);
> > > > > > > +
> > > > > > > +	/* Release the old block and set the out parameters. */
> > > > > > > +	xfs_btree_bload_drop_buf(cur->bc_tp, bpp);
> > > > > > > +	*blockp = new_block;
> > > > > > > +	*bpp = new_bp;
> > > > > > > +	xfs_btree_copy_ptrs(cur, ptrp, &new_ptr, 1);
> > > > > > > +	return 0;
> > > > > > > +}
> > > > > > > +
> > > > > > > +/* Load one leaf block. */
> > > > > > > +STATIC int
> > > > > > > +xfs_btree_bload_leaf(
> > > > > > > +	struct xfs_btree_cur		*cur,
> > > > > > > +	unsigned int			recs_this_block,
> > > > > > > +	xfs_btree_bload_get_fn		get_data,
> > > > > > > +	struct xfs_btree_block		*block,
> > > > > > > +	void				*priv)
> > > > > > > +{
> > > > > > > +	unsigned int			j;
> > > > > > > +	int				ret;
> > > > > > > +
> > > > > > > +	/* Fill the leaf block with records. */
> > > > > > > +	for (j = 1; j <= recs_this_block; j++) {
> > > > > > > +		union xfs_btree_rec	*block_recs;
> > > > > > > +
> > > > > > > +		ret = get_data(cur, priv);
> > > > > > > +		if (ret)
> > > > > > > +			return ret;
> > > > > > > +		block_recs = xfs_btree_rec_addr(cur, j, block);
> > > > > > > +		cur->bc_ops->init_rec_from_cur(cur, block_recs);
> > > > > > > +	}
> > > > > > > +
> > > > > > > +	return 0;
> > > > > > > +}
> > > > > > > +
> > > > > > > +/* Load one node block. */
> > > > > > > +STATIC int
> > > > > > > +xfs_btree_bload_node(
> > > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > > +	unsigned int		recs_this_block,
> > > > > > > +	union xfs_btree_ptr	*child_ptr,
> > > > > > > +	struct xfs_btree_block	*block)
> > > > > > > +{
> > > > > > > +	unsigned int		j;
> > > > > > > +	int			ret;
> > > > > > > +
> > > > > > > +	/* Fill the node block with keys and pointers. */
> > > > > > > +	for (j = 1; j <= recs_this_block; j++) {
> > > > > > > +		union xfs_btree_key	child_key;
> > > > > > > +		union xfs_btree_ptr	*block_ptr;
> > > > > > > +		union xfs_btree_key	*block_key;
> > > > > > > +		struct xfs_btree_block	*child_block;
> > > > > > > +		struct xfs_buf		*child_bp;
> > > > > > > +
> > > > > > > +		ASSERT(!xfs_btree_ptr_is_null(cur, child_ptr));
> > > > > > > +
> > > > > > > +		ret = xfs_btree_get_buf_block(cur, child_ptr, &child_block,
> > > > > > > +				&child_bp);
> > > > > > > +		if (ret)
> > > > > > > +			return ret;
> > > > > > > +
> > > > > > > +		xfs_btree_get_keys(cur, child_block, &child_key);
> > > > > > > +
> > > > > > > +		block_ptr = xfs_btree_ptr_addr(cur, j, block);
> > > > > > > +		xfs_btree_copy_ptrs(cur, block_ptr, child_ptr, 1);
> > > > > > > +
> > > > > > > +		block_key = xfs_btree_key_addr(cur, j, block);
> > > > > > > +		xfs_btree_copy_keys(cur, block_key, &child_key, 1);
> > > > > > > +
> > > > > > > +		xfs_btree_get_sibling(cur, child_block, child_ptr,
> > > > > > > +				XFS_BB_RIGHTSIB);
> > > > > > > +		xfs_trans_brelse(cur->bc_tp, child_bp);
> > > > > > > +	}
> > > > > > > +
> > > > > > > +	return 0;
> > > > > > > +}
> > > > > > > +
> > > > > > > +/*
> > > > > > > + * Compute the maximum number of records (or keyptrs) per block that we want to
> > > > > > > + * install at this level in the btree.  Caller is responsible for having set
> > > > > > > + * @cur->bc_private.b.forksize to the desired fork size, if appropriate.
> > > > > > > + */
> > > > > > > +STATIC unsigned int
> > > > > > > +xfs_btree_bload_max_npb(
> > > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > > +	struct xfs_btree_bload	*bbl,
> > > > > > > +	unsigned int		level)
> > > > > > > +{
> > > > > > > +	unsigned int		ret;
> > > > > > > +
> > > > > > > +	if (level == cur->bc_nlevels - 1 && cur->bc_ops->get_dmaxrecs)
> > > > > > > +		return cur->bc_ops->get_dmaxrecs(cur, level);
> > > > > > > +
> > > > > > > +	ret = cur->bc_ops->get_maxrecs(cur, level);
> > > > > > > +	if (level == 0)
> > > > > > > +		ret -= bbl->leaf_slack;
> > > > > > > +	else
> > > > > > > +		ret -= bbl->node_slack;
> > > > > > > +	return ret;
> > > > > > > +}
> > > > > > > +
> > > > > > > +/*
> > > > > > > + * Compute the desired number of records (or keyptrs) per block that we want to
> > > > > > > + * install at this level in the btree, which must be somewhere between minrecs
> > > > > > > + * and max_npb.  The caller is free to install fewer records per block.
> > > > > > > + */
> > > > > > > +STATIC unsigned int
> > > > > > > +xfs_btree_bload_desired_npb(
> > > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > > +	struct xfs_btree_bload	*bbl,
> > > > > > > +	unsigned int		level)
> > > > > > > +{
> > > > > > > +	unsigned int		npb = xfs_btree_bload_max_npb(cur, bbl, level);
> > > > > > > +
> > > > > > > +	/* Root blocks are not subject to minrecs rules. */
> > > > > > > +	if (level == cur->bc_nlevels - 1)
> > > > > > > +		return max(1U, npb);
> > > > > > > +
> > > > > > > +	return max_t(unsigned int, cur->bc_ops->get_minrecs(cur, level), npb);
> > > > > > > +}
> > > > > > > +
> > > > > > > +/*
> > > > > > > + * Compute the number of records to be stored in each block at this level and
> > > > > > > + * the number of blocks for this level.  For leaf levels, we must populate an
> > > > > > > + * empty root block even if there are no records, so we have to have at least
> > > > > > > + * one block.
> > > > > > > + */
> > > > > > > +STATIC void
> > > > > > > +xfs_btree_bload_level_geometry(
> > > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > > +	struct xfs_btree_bload	*bbl,
> > > > > > > +	unsigned int		level,
> > > > > > > +	uint64_t		nr_this_level,
> > > > > > > +	unsigned int		*avg_per_block,
> > > > > > > +	uint64_t		*blocks,
> > > > > > > +	uint64_t		*blocks_with_extra)
> > > > > > > +{
> > > > > > > +	uint64_t		npb;
> > > > > > > +	uint64_t		dontcare;
> > > > > > > +	unsigned int		desired_npb;
> > > > > > > +	unsigned int		maxnr;
> > > > > > > +
> > > > > > > +	maxnr = cur->bc_ops->get_maxrecs(cur, level);
> > > > > > > +
> > > > > > > +	/*
> > > > > > > +	 * Compute the number of blocks we need to fill each block with the
> > > > > > > +	 * desired number of records/keyptrs per block.  Because desired_npb
> > > > > > > +	 * could be minrecs, we use regular integer division (which rounds
> > > > > > > +	 * the block count down) so that in the next step the effective # of
> > > > > > > +	 * items per block will never be less than desired_npb.
> > > > > > > +	 */
> > > > > > > +	desired_npb = xfs_btree_bload_desired_npb(cur, bbl, level);
> > > > > > > +	*blocks = div64_u64_rem(nr_this_level, desired_npb, &dontcare);
> > > > > > > +	*blocks = max(1ULL, *blocks);
> > > > > > > +
> > > > > > > +	/*
> > > > > > > +	 * Compute the number of records that we will actually put in each
> > > > > > > +	 * block, assuming that we want to spread the records evenly between
> > > > > > > +	 * the blocks.  Take care that the effective # of items per block (npb)
> > > > > > > +	 * won't exceed maxrecs even for the blocks that get an extra record,
> > > > > > > +	 * since desired_npb could be maxrecs, and in the previous step we
> > > > > > > +	 * rounded the block count down.
> > > > > > > +	 */
> > > > > > > +	npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > > > > > +	if (npb > maxnr || (npb == maxnr && *blocks_with_extra > 0)) {
> > > > > > > +		(*blocks)++;
> > > > > > > +		npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
> > > > > > > +	}
> > > > > > > +
> > > > > > > +	*avg_per_block = min_t(uint64_t, npb, nr_this_level);
> > > > > > > +
> > > > > > > +	trace_xfs_btree_bload_level_geometry(cur, level, nr_this_level,
> > > > > > > +			*avg_per_block, desired_npb, *blocks,
> > > > > > > +			*blocks_with_extra);
> > > > > > > +}
> > > > > > > +
> > > > > > > +/*
> > > > > > > + * Ensure a slack value is appropriate for the btree.
> > > > > > > + *
> > > > > > > + * If the slack value is negative, set slack so that we fill the block to
> > > > > > > + * halfway between minrecs and maxrecs.  Make sure the slack is never so large
> > > > > > > + * that we can underflow minrecs.
> > > > > > > + */
> > > > > > > +static void
> > > > > > > +xfs_btree_bload_ensure_slack(
> > > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > > +	int			*slack,
> > > > > > > +	int			level)
> > > > > > > +{
> > > > > > > +	int			maxr;
> > > > > > > +	int			minr;
> > > > > > > +
> > > > > > > +	/*
> > > > > > > +	 * We only care about slack for btree blocks, so set the btree nlevels
> > > > > > > +	 * to 3 so that level 0 is a leaf block and level 1 is a node block.
> > > > > > > +	 * Avoid straying into inode roots, since we don't do slack there.
> > > > > > > +	 */
> > > > > > > +	cur->bc_nlevels = 3;
> > > > > > > +	maxr = cur->bc_ops->get_maxrecs(cur, level);
> > > > > > > +	minr = cur->bc_ops->get_minrecs(cur, level);
> > > > > > > +
> > > > > > > +	/*
> > > > > > > +	 * If slack is negative, automatically set slack so that we load the
> > > > > > > +	 * btree block approximately halfway between minrecs and maxrecs.
> > > > > > > +	 * Generally, this will net us 75% loading.
> > > > > > > +	 */
> > > > > > > +	if (*slack < 0)
> > > > > > > +		*slack = maxr - ((maxr + minr) >> 1);
> > > > > > > +
> > > > > > > +	*slack = min(*slack, maxr - minr);
> > > > > > > +}
> > > > > > > +
> > > > > > > +/*
> > > > > > > + * Prepare a btree cursor for a bulk load operation by computing the geometry
> > > > > > > + * fields in @bbl.  Caller must ensure that the btree cursor is a staging
> > > > > > > + * cursor.  This function can be called multiple times.
> > > > > > > + */
> > > > > > > +int
> > > > > > > +xfs_btree_bload_compute_geometry(
> > > > > > > +	struct xfs_btree_cur	*cur,
> > > > > > > +	struct xfs_btree_bload	*bbl,
> > > > > > > +	uint64_t		nr_records)
> > > > > > > +{
> > > > > > > +	uint64_t		nr_blocks = 0;
> > > > > > > +	uint64_t		nr_this_level;
> > > > > > > +
> > > > > > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > > > > > +
> > > > > > > +	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
> > > > > > > +	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
> > > > > > > +
> > > > > > > +	bbl->nr_records = nr_this_level = nr_records;
> > > > > > > +	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
> > > > > > > +		uint64_t	level_blocks;
> > > > > > > +		uint64_t	dontcare64;
> > > > > > > +		unsigned int	level = cur->bc_nlevels - 1;
> > > > > > > +		unsigned int	avg_per_block;
> > > > > > > +
> > > > > > > +		/*
> > > > > > > +		 * If all the things we want to store at this level would fit
> > > > > > > +		 * in a single root block, then we have our btree root and are
> > > > > > > +		 * done.  Note that bmap btrees do not allow records in the
> > > > > > > +		 * root.
> > > > > > > +		 */
> > > > > > > +		if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) || level != 0) {
> > > > > > > +			xfs_btree_bload_level_geometry(cur, bbl, level,
> > > > > > > +					nr_this_level, &avg_per_block,
> > > > > > > +					&level_blocks, &dontcare64);
> > > > > > > +			if (nr_this_level <= avg_per_block) {
> > > > > > > +				nr_blocks++;
> > > > > > > +				break;
> > > > > > > +			}
> > > > > > > +		}
> > > > > > > +
> > > > > > > +		/*
> > > > > > > +		 * Otherwise, we have to store all the records for this level
> > > > > > > +		 * in blocks and therefore need another level of btree to point
> > > > > > > +		 * to those blocks.  Increase the number of levels and
> > > > > > > +		 * recompute the number of records we can store at this level
> > > > > > > +		 * because that can change depending on whether or not a level
> > > > > > > +		 * is the root level.
> > > > > > > +		 */
> > > > > > > +		cur->bc_nlevels++;
> > > > > > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > > > > +				&avg_per_block, &level_blocks, &dontcare64);
> > > > > > > +		nr_blocks += level_blocks;
> > > > > > > +		nr_this_level = level_blocks;
> > > > > > > +	}
> > > > > > > +
> > > > > > > +	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
> > > > > > > +		return -EOVERFLOW;
> > > > > > > +
> > > > > > > +	bbl->btree_height = cur->bc_nlevels;
> > > > > > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
> > > > > > > +		bbl->nr_blocks = nr_blocks - 1;
> > > > > > > +	else
> > > > > > > +		bbl->nr_blocks = nr_blocks;
> > > > > > > +	return 0;
> > > > > > > +}
> > > > > > > +
> > > > > > > +/*
> > > > > > > + * Bulk load a btree.
> > > > > > > + *
> > > > > > > + * Load @bbl->nr_records quantity of records into a btree using the supplied
> > > > > > > + * empty and staging btree cursor @cur and a @bbl that has been filled out by
> > > > > > > + * the xfs_btree_bload_compute_geometry function.
> > > > > > > + *
> > > > > > > + * The @bbl->get_data function must populate the cursor's bc_rec every time it
> > > > > > > + * is called.  The @bbl->alloc_block function will be used to allocate new
> > > > > > > + * btree blocks.  @priv is passed to both functions.
> > > > > > > + *
> > > > > > > + * Caller must ensure that @cur is a staging cursor.  Any existing btree rooted
> > > > > > > + * in the fakeroot will be lost, so do not call this function twice.
> > > > > > > + */
> > > > > > > +int
> > > > > > > +xfs_btree_bload(
> > > > > > > +	struct xfs_btree_cur		*cur,
> > > > > > > +	struct xfs_btree_bload		*bbl,
> > > > > > > +	void				*priv)
> > > > > > > +{
> > > > > > > +	union xfs_btree_ptr		child_ptr;
> > > > > > > +	union xfs_btree_ptr		ptr;
> > > > > > > +	struct xfs_buf			*bp = NULL;
> > > > > > > +	struct xfs_btree_block		*block = NULL;
> > > > > > > +	uint64_t			nr_this_level = bbl->nr_records;
> > > > > > > +	uint64_t			blocks;
> > > > > > > +	uint64_t			i;
> > > > > > > +	uint64_t			blocks_with_extra;
> > > > > > > +	uint64_t			total_blocks = 0;
> > > > > > > +	unsigned int			avg_per_block;
> > > > > > > +	unsigned int			level = 0;
> > > > > > > +	int				ret;
> > > > > > > +
> > > > > > > +	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
> > > > > > > +
> > > > > > > +	cur->bc_nlevels = bbl->btree_height;
> > > > > > > +	xfs_btree_set_ptr_null(cur, &child_ptr);
> > > > > > > +	xfs_btree_set_ptr_null(cur, &ptr);
> > > > > > > +
> > > > > > > +	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > > > > +			&avg_per_block, &blocks, &blocks_with_extra);
> > > > > > > +
> > > > > > > +	/* Load each leaf block. */
> > > > > > > +	for (i = 0; i < blocks; i++) {
> > > > > > > +		unsigned int		nr_this_block = avg_per_block;
> > > > > > > +
> > > > > > > +		if (i < blocks_with_extra)
> > > > > > > +			nr_this_block++;
> > > > > > > +
> > > > > > > +		ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > > > > > +				nr_this_block, &ptr, &bp, &block, priv);
> > > > > > > +		if (ret)
> > > > > > > +			return ret;
> > > > > > > +
> > > > > > > +		trace_xfs_btree_bload_block(cur, level, i, blocks, &ptr,
> > > > > > > +				nr_this_block);
> > > > > > > +
> > > > > > > +		ret = xfs_btree_bload_leaf(cur, nr_this_block, bbl->get_data,
> > > > > > > +				block, priv);
> > > > > > > +		if (ret)
> > > > > > > +			goto out;
> > > > > > > +
> > > > > > > +		/* Record the leftmost pointer to start the next level. */
> > > > > > > +		if (i == 0)
> > > > > > > +			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
> > > > > > > +	}
> > > > > > > +	total_blocks += blocks;
> > > > > > > +	xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > > > > > +
> > > > > > > +	/* Populate the internal btree nodes. */
> > > > > > > +	for (level = 1; level < cur->bc_nlevels; level++) {
> > > > > > > +		union xfs_btree_ptr	first_ptr;
> > > > > > > +
> > > > > > > +		nr_this_level = blocks;
> > > > > > > +		block = NULL;
> > > > > > > +		xfs_btree_set_ptr_null(cur, &ptr);
> > > > > > > +
> > > > > > > +		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
> > > > > > > +				&avg_per_block, &blocks, &blocks_with_extra);
> > > > > > > +
> > > > > > > +		/* Load each node block. */
> > > > > > > +		for (i = 0; i < blocks; i++) {
> > > > > > > +			unsigned int	nr_this_block = avg_per_block;
> > > > > > > +
> > > > > > > +			if (i < blocks_with_extra)
> > > > > > > +				nr_this_block++;
> > > > > > > +
> > > > > > > +			ret = xfs_btree_bload_prep_block(cur, bbl, level,
> > > > > > > +					nr_this_block, &ptr, &bp, &block,
> > > > > > > +					priv);
> > > > > > > +			if (ret)
> > > > > > > +				return ret;
> > > > > > > +
> > > > > > > +			trace_xfs_btree_bload_block(cur, level, i, blocks,
> > > > > > > +					&ptr, nr_this_block);
> > > > > > > +
> > > > > > > +			ret = xfs_btree_bload_node(cur, nr_this_block,
> > > > > > > +					&child_ptr, block);
> > > > > > > +			if (ret)
> > > > > > > +				goto out;
> > > > > > > +
> > > > > > > +			/*
> > > > > > > +			 * Record the leftmost pointer to start the next level.
> > > > > > > +			 */
> > > > > > > +			if (i == 0)
> > > > > > > +				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
> > > > > > > +		}
> > > > > > > +		total_blocks += blocks;
> > > > > > > +		xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
> > > > > > > +		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
> > > > > > > +	}
> > > > > > > +
> > > > > > > +	/* Initialize the new root. */
> > > > > > > +	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
> > > > > > > +		ASSERT(xfs_btree_ptr_is_null(cur, &ptr));
> > > > > > > +		cur->bc_private.b.ifake->if_levels = cur->bc_nlevels;
> > > > > > > +		cur->bc_private.b.ifake->if_blocks = total_blocks - 1;
> > > > > > > +	} else {
> > > > > > > +		cur->bc_private.a.afake->af_root = be32_to_cpu(ptr.s);
> > > > > > > +		cur->bc_private.a.afake->af_levels = cur->bc_nlevels;
> > > > > > > +		cur->bc_private.a.afake->af_blocks = total_blocks;
> > > > > > > +	}
> > > > > > > +out:
> > > > > > > +	if (bp)
> > > > > > > +		xfs_trans_brelse(cur->bc_tp, bp);
> > > > > > > +	return ret;
> > > > > > > +}
> > > > > > > diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
> > > > > > > index a17becb72ab8..5c6992a04ea2 100644
> > > > > > > --- a/fs/xfs/libxfs/xfs_btree.h
> > > > > > > +++ b/fs/xfs/libxfs/xfs_btree.h
> > > > > > > @@ -582,4 +582,47 @@ void xfs_btree_stage_ifakeroot(struct xfs_btree_cur *cur,
> > > > > > >  void xfs_btree_commit_ifakeroot(struct xfs_btree_cur *cur, int whichfork,
> > > > > > >  		const struct xfs_btree_ops *ops);
> > > > > > >  
> > > > > > > +typedef int (*xfs_btree_bload_get_fn)(struct xfs_btree_cur *cur, void *priv);
> > > > > > > +typedef int (*xfs_btree_bload_alloc_block_fn)(struct xfs_btree_cur *cur,
> > > > > > > +		union xfs_btree_ptr *ptr, void *priv);
> > > > > > > +typedef size_t (*xfs_btree_bload_iroot_size_fn)(struct xfs_btree_cur *cur,
> > > > > > > +		unsigned int nr_this_level, void *priv);
> > > > > > > +
> > > > > > > +/* Bulk loading of staged btrees. */
> > > > > > > +struct xfs_btree_bload {
> > > > > > > +	/* Function to store a record in the cursor. */
> > > > > > > +	xfs_btree_bload_get_fn		get_data;
> > > > > > > +
> > > > > > > +	/* Function to allocate a block for the btree. */
> > > > > > > +	xfs_btree_bload_alloc_block_fn	alloc_block;
> > > > > > > +
> > > > > > > +	/* Function to compute the size of the in-core btree root block. */
> > > > > > > +	xfs_btree_bload_iroot_size_fn	iroot_size;
> > > > > > > +
> > > > > > > +	/* Number of records the caller wants to store. */
> > > > > > > +	uint64_t			nr_records;
> > > > > > > +
> > > > > > > +	/* Number of btree blocks needed to store those records. */
> > > > > > > +	uint64_t			nr_blocks;
> > > > > > > +
> > > > > > > +	/*
> > > > > > > +	 * Number of free records to leave in each leaf block.  If this (or
> > > > > > > +	 * any of the slack values) are negative, this will be computed to
> > > > > > > +	 * be halfway between maxrecs and minrecs.  This typically leaves the
> > > > > > > +	 * block 75% full.
> > > > > > > +	 */
> > > > > > > +	int				leaf_slack;
> > > > > > > +
> > > > > > > +	/* Number of free keyptrs to leave in each node block. */
> > > > > > > +	int				node_slack;
> > > > > > > +
> > > > > > > +	/* Computed btree height. */
> > > > > > > +	unsigned int			btree_height;
> > > > > > > +};
> > > > > > > +
> > > > > > > +int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
> > > > > > > +		struct xfs_btree_bload *bbl, uint64_t nr_records);
> > > > > > > +int xfs_btree_bload(struct xfs_btree_cur *cur, struct xfs_btree_bload *bbl,
> > > > > > > +		void *priv);
> > > > > > > +
> > > > > > >  #endif	/* __XFS_BTREE_H__ */
> > > > > > > diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
> > > > > > > index bc85b89f88ca..9b5e58a92381 100644
> > > > > > > --- a/fs/xfs/xfs_trace.c
> > > > > > > +++ b/fs/xfs/xfs_trace.c
> > > > > > > @@ -6,6 +6,7 @@
> > > > > > >  #include "xfs.h"
> > > > > > >  #include "xfs_fs.h"
> > > > > > >  #include "xfs_shared.h"
> > > > > > > +#include "xfs_bit.h"
> > > > > > >  #include "xfs_format.h"
> > > > > > >  #include "xfs_log_format.h"
> > > > > > >  #include "xfs_trans_resv.h"
> > > > > > > diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
> > > > > > > index a78055521fcd..6d7ba64b7a0f 100644
> > > > > > > --- a/fs/xfs/xfs_trace.h
> > > > > > > +++ b/fs/xfs/xfs_trace.h
> > > > > > > @@ -35,6 +35,7 @@ struct xfs_icreate_log;
> > > > > > >  struct xfs_owner_info;
> > > > > > >  struct xfs_trans_res;
> > > > > > >  struct xfs_inobt_rec_incore;
> > > > > > > +union xfs_btree_ptr;
> > > > > > >  
> > > > > > >  DECLARE_EVENT_CLASS(xfs_attr_list_class,
> > > > > > >  	TP_PROTO(struct xfs_attr_list_context *ctx),
> > > > > > > @@ -3670,6 +3671,90 @@ TRACE_EVENT(xfs_btree_commit_ifakeroot,
> > > > > > >  		  __entry->blocks)
> > > > > > >  )
> > > > > > >  
> > > > > > > +TRACE_EVENT(xfs_btree_bload_level_geometry,
> > > > > > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > > > > > +		 uint64_t nr_this_level, unsigned int nr_per_block,
> > > > > > > +		 unsigned int desired_npb, uint64_t blocks,
> > > > > > > +		 uint64_t blocks_with_extra),
> > > > > > > +	TP_ARGS(cur, level, nr_this_level, nr_per_block, desired_npb, blocks,
> > > > > > > +		blocks_with_extra),
> > > > > > > +	TP_STRUCT__entry(
> > > > > > > +		__field(dev_t, dev)
> > > > > > > +		__field(xfs_btnum_t, btnum)
> > > > > > > +		__field(unsigned int, level)
> > > > > > > +		__field(unsigned int, nlevels)
> > > > > > > +		__field(uint64_t, nr_this_level)
> > > > > > > +		__field(unsigned int, nr_per_block)
> > > > > > > +		__field(unsigned int, desired_npb)
> > > > > > > +		__field(unsigned long long, blocks)
> > > > > > > +		__field(unsigned long long, blocks_with_extra)
> > > > > > > +	),
> > > > > > > +	TP_fast_assign(
> > > > > > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > > > > > +		__entry->btnum = cur->bc_btnum;
> > > > > > > +		__entry->level = level;
> > > > > > > +		__entry->nlevels = cur->bc_nlevels;
> > > > > > > +		__entry->nr_this_level = nr_this_level;
> > > > > > > +		__entry->nr_per_block = nr_per_block;
> > > > > > > +		__entry->desired_npb = desired_npb;
> > > > > > > +		__entry->blocks = blocks;
> > > > > > > +		__entry->blocks_with_extra = blocks_with_extra;
> > > > > > > +	),
> > > > > > > +	TP_printk("dev %d:%d btree %s level %u/%u nr_this_level %llu nr_per_block %u desired_npb %u blocks %llu blocks_with_extra %llu",
> > > > > > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > > > > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > > > > > +		  __entry->level,
> > > > > > > +		  __entry->nlevels,
> > > > > > > +		  __entry->nr_this_level,
> > > > > > > +		  __entry->nr_per_block,
> > > > > > > +		  __entry->desired_npb,
> > > > > > > +		  __entry->blocks,
> > > > > > > +		  __entry->blocks_with_extra)
> > > > > > > +)
> > > > > > > +
> > > > > > > +TRACE_EVENT(xfs_btree_bload_block,
> > > > > > > +	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
> > > > > > > +		 uint64_t block_idx, uint64_t nr_blocks,
> > > > > > > +		 union xfs_btree_ptr *ptr, unsigned int nr_records),
> > > > > > > +	TP_ARGS(cur, level, block_idx, nr_blocks, ptr, nr_records),
> > > > > > > +	TP_STRUCT__entry(
> > > > > > > +		__field(dev_t, dev)
> > > > > > > +		__field(xfs_btnum_t, btnum)
> > > > > > > +		__field(unsigned int, level)
> > > > > > > +		__field(unsigned long long, block_idx)
> > > > > > > +		__field(unsigned long long, nr_blocks)
> > > > > > > +		__field(xfs_agnumber_t, agno)
> > > > > > > +		__field(xfs_agblock_t, agbno)
> > > > > > > +		__field(unsigned int, nr_records)
> > > > > > > +	),
> > > > > > > +	TP_fast_assign(
> > > > > > > +		__entry->dev = cur->bc_mp->m_super->s_dev;
> > > > > > > +		__entry->btnum = cur->bc_btnum;
> > > > > > > +		__entry->level = level;
> > > > > > > +		__entry->block_idx = block_idx;
> > > > > > > +		__entry->nr_blocks = nr_blocks;
> > > > > > > +		if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
> > > > > > > +			xfs_fsblock_t	fsb = be64_to_cpu(ptr->l);
> > > > > > > +
> > > > > > > +			__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsb);
> > > > > > > +			__entry->agbno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsb);
> > > > > > > +		} else {
> > > > > > > +			__entry->agno = cur->bc_private.a.agno;
> > > > > > > +			__entry->agbno = be32_to_cpu(ptr->s);
> > > > > > > +		}
> > > > > > > +		__entry->nr_records = nr_records;
> > > > > > > +	),
> > > > > > > +	TP_printk("dev %d:%d btree %s level %u block %llu/%llu fsb (%u/%u) recs %u",
> > > > > > > +		  MAJOR(__entry->dev), MINOR(__entry->dev),
> > > > > > > +		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
> > > > > > > +		  __entry->level,
> > > > > > > +		  __entry->block_idx,
> > > > > > > +		  __entry->nr_blocks,
> > > > > > > +		  __entry->agno,
> > > > > > > +		  __entry->agbno,
> > > > > > > +		  __entry->nr_records)
> > > > > > > +)
> > > > > > > +
> > > > > > >  #endif /* _TRACE_XFS_H */
> > > > > > >  
> > > > > > >  #undef TRACE_INCLUDE_PATH
> > > > > > >

Patch

diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
index 4b06d5d86834..17b0fdb87729 100644
--- a/fs/xfs/libxfs/xfs_btree.c
+++ b/fs/xfs/libxfs/xfs_btree.c
@@ -1382,7 +1382,7 @@  STATIC void
 xfs_btree_copy_ptrs(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*dst_ptr,
-	union xfs_btree_ptr	*src_ptr,
+	const union xfs_btree_ptr *src_ptr,
 	int			numptrs)
 {
 	ASSERT(numptrs >= 0);
@@ -5104,3 +5104,567 @@  xfs_btree_commit_ifakeroot(
 	cur->bc_ops = ops;
 	cur->bc_flags &= ~XFS_BTREE_STAGING;
 }
+
+/*
+ * Bulk Loading of Staged Btrees
+ * =============================
+ *
+ * This interface is used with a staged btree cursor to create a totally new
+ * btree with a large number of records (i.e. more than what would fit in a
+ * single block).  When the creation is complete, the new root can be linked
+ * atomically into the filesystem by committing the staged cursor.
+ *
+ * The first step for the caller is to construct a fake btree root structure
+ * and a staged btree cursor.  A staging cursor contains all the geometry
+ * information for the btree type but will fail all operations that could have
+ * side effects in the filesystem (e.g. btree shape changes).  Regular
+ * operations will not work unless the staging cursor is committed and becomes
+ * a regular cursor.
+ *
+ * For a btree rooted in an AG header, use an xbtree_afakeroot structure.
+ * This should be initialized to zero.  For a btree rooted in an inode fork,
+ * use an xbtree_ifakeroot structure.  @if_fork_size field should be set to
+ * the number of bytes available to the fork in the inode; @if_fork should
+ * point to a freshly allocated xfs_inode_fork; and @if_format should be set
+ * to the appropriate fork type (e.g. XFS_DINODE_FMT_BTREE).
+ *
+ * The next step for the caller is to initialize a struct xfs_btree_bload
+ * context.  The @nr_records field is the number of records that are to be
+ * loaded into the btree.  The @leaf_slack and @node_slack fields are the
+ * number of records (or key/ptr) slots to leave empty in new btree blocks.
+ * If a caller sets a slack value to -1, the slack value will be computed to
+ * fill the block halfway between minrecs and maxrecs items per block.
+ *
+ * The number of items placed in each btree block is computed via the following
+ * algorithm: For leaf levels, the number of items for the level is nr_records.
+ * For node levels, the number of items for the level is the number of blocks
+ * in the next lower level of the tree.  For each level, the desired number of
+ * items per block is defined as:
+ *
+ * desired = max(minrecs, maxrecs - slack factor)
+ *
+ * The number of blocks for the level is defined to be:
+ *
+ * blocks = nr_items / desired
+ *
+ * Note this is rounded down so that the npb calculation below will never fall
+ * below minrecs.  The number of items that will actually be loaded into each
+ * btree block is defined as:
+ *
+ * npb =  nr_items / blocks
+ *
+ * Some of the leftmost blocks in the level will contain one extra record as
+ * needed to handle uneven division.  If the number of records in any block
+ * would exceed maxrecs for that level, blocks is incremented and npb is
+ * recalculated.
+ *
+ * In other words, we compute the number of blocks needed to satisfy a given
+ * loading level, then spread the items as evenly as possible.
+ *
+ * To complete this step, call xfs_btree_bload_compute_geometry, which uses
+ * those settings to compute the height of the btree and the number of blocks
+ * that will be needed to construct the btree.  These values are stored in the
+ * @btree_height and @nr_blocks fields.
+ *
+ * At this point, the caller must allocate @nr_blocks blocks and save them for
+ * later.  If space is to be allocated transactionally, the staging cursor
+ * must be deleted before and recreated after, which is why computing the
+ * geometry is a separate step.
+ *
+ * The fourth step in the bulk loading process is to set the function pointers
+ * in the bload context structure.  @get_data will be called for each record
+ * that will be loaded into the btree; it should set the cursor's bc_rec
+ * field, which will be converted to on-disk format and copied into the
+ * appropriate record slot.  @alloc_block should supply one of the blocks
+ * allocated in the previous step.  For btrees which are rooted in an inode
+ * fork, @iroot_size is called to compute the size of the incore btree root
+ * block.  Call xfs_btree_bload to start constructing the btree.
+ *
+ * The final step is to commit the staging cursor, which logs the new btree
+ * root and turns the btree into a regular btree cursor, and free the fake
+ * roots.
+ */
+
+/*
+ * Put a btree block that we're loading onto the ordered list and release it.
+ * The btree blocks will be written when the final transaction swapping the
+ * btree roots is committed.
+ */
+static void
+xfs_btree_bload_drop_buf(
+	struct xfs_trans	*tp,
+	struct xfs_buf		**bpp)
+{
+	if (*bpp == NULL)
+		return;
+
+	xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_BTREE_BUF);
+	xfs_trans_ordered_buf(tp, *bpp);
+	xfs_trans_brelse(tp, *bpp);
+	*bpp = NULL;
+}
+
+/* Allocate and initialize one btree block for bulk loading. */
+STATIC int
+xfs_btree_bload_prep_block(
+	struct xfs_btree_cur		*cur,
+	struct xfs_btree_bload		*bbl,
+	unsigned int			level,
+	unsigned int			nr_this_block,
+	union xfs_btree_ptr		*ptrp,
+	struct xfs_buf			**bpp,
+	struct xfs_btree_block		**blockp,
+	void				*priv)
+{
+	union xfs_btree_ptr		new_ptr;
+	struct xfs_buf			*new_bp;
+	struct xfs_btree_block		*new_block;
+	int				ret;
+
+	if ((cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) &&
+	    level == cur->bc_nlevels - 1) {
+		struct xfs_ifork	*ifp = cur->bc_private.b.ifake->if_fork;
+		size_t			new_size;
+
+		/* Allocate a new incore btree root block. */
+		new_size = bbl->iroot_size(cur, nr_this_block, priv);
+		ifp->if_broot = kmem_zalloc(new_size, 0);
+		ifp->if_broot_bytes = (int)new_size;
+		ifp->if_flags |= XFS_IFBROOT;
+
+		/* Initialize it and send it out. */
+		xfs_btree_init_block_int(cur->bc_mp, ifp->if_broot,
+				XFS_BUF_DADDR_NULL, cur->bc_btnum, level,
+				nr_this_block, cur->bc_private.b.ip->i_ino,
+				cur->bc_flags);
+
+		*bpp = NULL;
+		*blockp = ifp->if_broot;
+		xfs_btree_set_ptr_null(cur, ptrp);
+		return 0;
+	}
+
+	/* Allocate a new leaf block. */
+	ret = bbl->alloc_block(cur, &new_ptr, priv);
+	if (ret)
+		return ret;
+
+	ASSERT(!xfs_btree_ptr_is_null(cur, &new_ptr));
+
+	ret = xfs_btree_get_buf_block(cur, &new_ptr, &new_block, &new_bp);
+	if (ret)
+		return ret;
+
+	/* Initialize the btree block. */
+	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
+	if (*blockp)
+		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
+	xfs_btree_set_sibling(cur, new_block, ptrp, XFS_BB_LEFTSIB);
+	xfs_btree_set_numrecs(new_block, nr_this_block);
+
+	/* Release the old block and set the out parameters. */
+	xfs_btree_bload_drop_buf(cur->bc_tp, bpp);
+	*blockp = new_block;
+	*bpp = new_bp;
+	xfs_btree_copy_ptrs(cur, ptrp, &new_ptr, 1);
+	return 0;
+}
+
+/* Load one leaf block. */
+STATIC int
+xfs_btree_bload_leaf(
+	struct xfs_btree_cur		*cur,
+	unsigned int			recs_this_block,
+	xfs_btree_bload_get_fn		get_data,
+	struct xfs_btree_block		*block,
+	void				*priv)
+{
+	unsigned int			j;
+	int				ret;
+
+	/* Fill the leaf block with records. */
+	for (j = 1; j <= recs_this_block; j++) {
+		union xfs_btree_rec	*block_recs;
+
+		ret = get_data(cur, priv);
+		if (ret)
+			return ret;
+		block_recs = xfs_btree_rec_addr(cur, j, block);
+		cur->bc_ops->init_rec_from_cur(cur, block_recs);
+	}
+
+	return 0;
+}
+
+/* Load one node block. */
+STATIC int
+xfs_btree_bload_node(
+	struct xfs_btree_cur	*cur,
+	unsigned int		recs_this_block,
+	union xfs_btree_ptr	*child_ptr,
+	struct xfs_btree_block	*block)
+{
+	unsigned int		j;
+	int			ret;
+
+	/* Fill the node block with keys and pointers. */
+	for (j = 1; j <= recs_this_block; j++) {
+		union xfs_btree_key	child_key;
+		union xfs_btree_ptr	*block_ptr;
+		union xfs_btree_key	*block_key;
+		struct xfs_btree_block	*child_block;
+		struct xfs_buf		*child_bp;
+
+		ASSERT(!xfs_btree_ptr_is_null(cur, child_ptr));
+
+		ret = xfs_btree_get_buf_block(cur, child_ptr, &child_block,
+				&child_bp);
+		if (ret)
+			return ret;
+
+		xfs_btree_get_keys(cur, child_block, &child_key);
+
+		block_ptr = xfs_btree_ptr_addr(cur, j, block);
+		xfs_btree_copy_ptrs(cur, block_ptr, child_ptr, 1);
+
+		block_key = xfs_btree_key_addr(cur, j, block);
+		xfs_btree_copy_keys(cur, block_key, &child_key, 1);
+
+		xfs_btree_get_sibling(cur, child_block, child_ptr,
+				XFS_BB_RIGHTSIB);
+		xfs_trans_brelse(cur->bc_tp, child_bp);
+	}
+
+	return 0;
+}
+
+/*
+ * Compute the maximum number of records (or keyptrs) per block that we want to
+ * install at this level in the btree.  Caller is responsible for having set
+ * @cur->bc_private.b.forksize to the desired fork size, if appropriate.
+ */
+STATIC unsigned int
+xfs_btree_bload_max_npb(
+	struct xfs_btree_cur	*cur,
+	struct xfs_btree_bload	*bbl,
+	unsigned int		level)
+{
+	unsigned int		ret;
+
+	if (level == cur->bc_nlevels - 1 && cur->bc_ops->get_dmaxrecs)
+		return cur->bc_ops->get_dmaxrecs(cur, level);
+
+	ret = cur->bc_ops->get_maxrecs(cur, level);
+	if (level == 0)
+		ret -= bbl->leaf_slack;
+	else
+		ret -= bbl->node_slack;
+	return ret;
+}
+
+/*
+ * Compute the desired number of records (or keyptrs) per block that we want to
+ * install at this level in the btree, which must be somewhere between minrecs
+ * and max_npb.  The caller is free to install fewer records per block.
+ */
+STATIC unsigned int
+xfs_btree_bload_desired_npb(
+	struct xfs_btree_cur	*cur,
+	struct xfs_btree_bload	*bbl,
+	unsigned int		level)
+{
+	unsigned int		npb = xfs_btree_bload_max_npb(cur, bbl, level);
+
+	/* Root blocks are not subject to minrecs rules. */
+	if (level == cur->bc_nlevels - 1)
+		return max(1U, npb);
+
+	return max_t(unsigned int, cur->bc_ops->get_minrecs(cur, level), npb);
+}
+
+/*
+ * Compute the number of records to be stored in each block at this level and
+ * the number of blocks for this level.  For leaf levels, we must populate an
+ * empty root block even if there are no records, so we have to have at least
+ * one block.
+ */
+STATIC void
+xfs_btree_bload_level_geometry(
+	struct xfs_btree_cur	*cur,
+	struct xfs_btree_bload	*bbl,
+	unsigned int		level,
+	uint64_t		nr_this_level,
+	unsigned int		*avg_per_block,
+	uint64_t		*blocks,
+	uint64_t		*blocks_with_extra)
+{
+	uint64_t		npb;
+	uint64_t		dontcare;
+	unsigned int		desired_npb;
+	unsigned int		maxnr;
+
+	maxnr = cur->bc_ops->get_maxrecs(cur, level);
+
+	/*
+	 * Compute the number of blocks we need to fill each block with the
+	 * desired number of records/keyptrs per block.  Because desired_npb
+	 * could be minrecs, we use regular integer division (which rounds
+	 * the block count down) so that in the next step the effective # of
+	 * items per block will never be less than desired_npb.
+	 */
+	desired_npb = xfs_btree_bload_desired_npb(cur, bbl, level);
+	*blocks = div64_u64_rem(nr_this_level, desired_npb, &dontcare);
+	*blocks = max(1ULL, *blocks);
+
+	/*
+	 * Compute the number of records that we will actually put in each
+	 * block, assuming that we want to spread the records evenly between
+	 * the blocks.  Take care that the effective # of items per block (npb)
+	 * won't exceed maxrecs even for the blocks that get an extra record,
+	 * since desired_npb could be maxrecs, and in the previous step we
+	 * rounded the block count down.
+	 */
+	npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
+	if (npb > maxnr || (npb == maxnr && *blocks_with_extra > 0)) {
+		(*blocks)++;
+		npb = div64_u64_rem(nr_this_level, *blocks, blocks_with_extra);
+	}
+
+	*avg_per_block = min_t(uint64_t, npb, nr_this_level);
+
+	trace_xfs_btree_bload_level_geometry(cur, level, nr_this_level,
+			*avg_per_block, desired_npb, *blocks,
+			*blocks_with_extra);
+}
+
+/*
+ * Ensure a slack value is appropriate for the btree.
+ *
+ * If the slack value is negative, set slack so that we fill the block to
+ * halfway between minrecs and maxrecs.  Make sure the slack is never so large
+ * that we can underflow minrecs.
+ */
+static void
+xfs_btree_bload_ensure_slack(
+	struct xfs_btree_cur	*cur,
+	int			*slack,
+	int			level)
+{
+	int			maxr;
+	int			minr;
+
+	/*
+	 * We only care about slack for btree blocks, so set the btree nlevels
+	 * to 3 so that level 0 is a leaf block and level 1 is a node block.
+	 * Avoid straying into inode roots, since we don't do slack there.
+	 */
+	cur->bc_nlevels = 3;
+	maxr = cur->bc_ops->get_maxrecs(cur, level);
+	minr = cur->bc_ops->get_minrecs(cur, level);
+
+	/*
+	 * If slack is negative, automatically set slack so that we load the
+	 * btree block approximately halfway between minrecs and maxrecs.
+	 * Generally, this will net us 75% loading.
+	 */
+	if (*slack < 0)
+		*slack = maxr - ((maxr + minr) >> 1);
+
+	*slack = min(*slack, maxr - minr);
+}
+
+/*
+ * Prepare a btree cursor for a bulk load operation by computing the geometry
+ * fields in @bbl.  Caller must ensure that the btree cursor is a staging
+ * cursor.  This function can be called multiple times.
+ */
+int
+xfs_btree_bload_compute_geometry(
+	struct xfs_btree_cur	*cur,
+	struct xfs_btree_bload	*bbl,
+	uint64_t		nr_records)
+{
+	uint64_t		nr_blocks = 0;
+	uint64_t		nr_this_level;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+
+	xfs_btree_bload_ensure_slack(cur, &bbl->leaf_slack, 0);
+	xfs_btree_bload_ensure_slack(cur, &bbl->node_slack, 1);
+
+	bbl->nr_records = nr_this_level = nr_records;
+	for (cur->bc_nlevels = 1; cur->bc_nlevels < XFS_BTREE_MAXLEVELS;) {
+		uint64_t	level_blocks;
+		uint64_t	dontcare64;
+		unsigned int	level = cur->bc_nlevels - 1;
+		unsigned int	avg_per_block;
+
+		/*
+		 * If all the things we want to store at this level would fit
+		 * in a single root block, then we have our btree root and are
+		 * done.  Note that bmap btrees do not allow records in the
+		 * root.
+		 */
+		if (!(cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) || level != 0) {
+			xfs_btree_bload_level_geometry(cur, bbl, level,
+					nr_this_level, &avg_per_block,
+					&level_blocks, &dontcare64);
+			if (nr_this_level <= avg_per_block) {
+				nr_blocks++;
+				break;
+			}
+		}
+
+		/*
+		 * Otherwise, we have to store all the records for this level
+		 * in blocks and therefore need another level of btree to point
+		 * to those blocks.  Increase the number of levels and
+		 * recompute the number of records we can store at this level
+		 * because that can change depending on whether or not a level
+		 * is the root level.
+		 */
+		cur->bc_nlevels++;
+		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
+				&avg_per_block, &level_blocks, &dontcare64);
+		nr_blocks += level_blocks;
+		nr_this_level = level_blocks;
+	}
+
+	if (cur->bc_nlevels == XFS_BTREE_MAXLEVELS)
+		return -EOVERFLOW;
+
+	bbl->btree_height = cur->bc_nlevels;
+	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
+		bbl->nr_blocks = nr_blocks - 1;
+	else
+		bbl->nr_blocks = nr_blocks;
+	return 0;
+}
+
+/*
+ * Bulk load a btree.
+ *
+ * Load @bbl->nr_records quantity of records into a btree using the supplied
+ * empty and staging btree cursor @cur and a @bbl that has been filled out by
+ * the xfs_btree_bload_compute_geometry function.
+ *
+ * The @bbl->get_data function must populate the cursor's bc_rec every time it
+ * is called.  The @bbl->alloc_block function will be used to allocate new
+ * btree blocks.  @priv is passed to both functions.
+ *
+ * Caller must ensure that @cur is a staging cursor.  Any existing btree rooted
+ * in the fakeroot will be lost, so do not call this function twice.
+ */
+int
+xfs_btree_bload(
+	struct xfs_btree_cur		*cur,
+	struct xfs_btree_bload		*bbl,
+	void				*priv)
+{
+	union xfs_btree_ptr		child_ptr;
+	union xfs_btree_ptr		ptr;
+	struct xfs_buf			*bp = NULL;
+	struct xfs_btree_block		*block = NULL;
+	uint64_t			nr_this_level = bbl->nr_records;
+	uint64_t			blocks;
+	uint64_t			i;
+	uint64_t			blocks_with_extra;
+	uint64_t			total_blocks = 0;
+	unsigned int			avg_per_block;
+	unsigned int			level = 0;
+	int				ret;
+
+	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
+
+	cur->bc_nlevels = bbl->btree_height;
+	xfs_btree_set_ptr_null(cur, &child_ptr);
+	xfs_btree_set_ptr_null(cur, &ptr);
+
+	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
+			&avg_per_block, &blocks, &blocks_with_extra);
+
+	/* Load each leaf block. */
+	for (i = 0; i < blocks; i++) {
+		unsigned int		nr_this_block = avg_per_block;
+
+		if (i < blocks_with_extra)
+			nr_this_block++;
+
+		ret = xfs_btree_bload_prep_block(cur, bbl, level,
+				nr_this_block, &ptr, &bp, &block, priv);
+		if (ret)
+			return ret;
+
+		trace_xfs_btree_bload_block(cur, level, i, blocks, &ptr,
+				nr_this_block);
+
+		ret = xfs_btree_bload_leaf(cur, nr_this_block, bbl->get_data,
+				block, priv);
+		if (ret)
+			goto out;
+
+		/* Record the leftmost pointer to start the next level. */
+		if (i == 0)
+			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
+	}
+	total_blocks += blocks;
+	xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
+
+	/* Populate the internal btree nodes. */
+	for (level = 1; level < cur->bc_nlevels; level++) {
+		union xfs_btree_ptr	first_ptr;
+
+		nr_this_level = blocks;
+		block = NULL;
+		xfs_btree_set_ptr_null(cur, &ptr);
+
+		xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
+				&avg_per_block, &blocks, &blocks_with_extra);
+
+		/* Load each node block. */
+		for (i = 0; i < blocks; i++) {
+			unsigned int	nr_this_block = avg_per_block;
+
+			if (i < blocks_with_extra)
+				nr_this_block++;
+
+			ret = xfs_btree_bload_prep_block(cur, bbl, level,
+					nr_this_block, &ptr, &bp, &block,
+					priv);
+			if (ret)
+				return ret;
+
+			trace_xfs_btree_bload_block(cur, level, i, blocks,
+					&ptr, nr_this_block);
+
+			ret = xfs_btree_bload_node(cur, nr_this_block,
+					&child_ptr, block);
+			if (ret)
+				goto out;
+
+			/*
+			 * Record the leftmost pointer to start the next level.
+			 */
+			if (i == 0)
+				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
+		}
+		total_blocks += blocks;
+		xfs_btree_bload_drop_buf(cur->bc_tp, &bp);
+		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
+	}
+
+	/* Initialize the new root. */
+	if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE) {
+		ASSERT(xfs_btree_ptr_is_null(cur, &ptr));
+		cur->bc_private.b.ifake->if_levels = cur->bc_nlevels;
+		cur->bc_private.b.ifake->if_blocks = total_blocks - 1;
+	} else {
+		cur->bc_private.a.afake->af_root = be32_to_cpu(ptr.s);
+		cur->bc_private.a.afake->af_levels = cur->bc_nlevels;
+		cur->bc_private.a.afake->af_blocks = total_blocks;
+	}
+out:
+	if (bp)
+		xfs_trans_brelse(cur->bc_tp, bp);
+	return ret;
+}
diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
index a17becb72ab8..5c6992a04ea2 100644
--- a/fs/xfs/libxfs/xfs_btree.h
+++ b/fs/xfs/libxfs/xfs_btree.h
@@ -582,4 +582,47 @@  void xfs_btree_stage_ifakeroot(struct xfs_btree_cur *cur,
 void xfs_btree_commit_ifakeroot(struct xfs_btree_cur *cur, int whichfork,
 		const struct xfs_btree_ops *ops);
 
+typedef int (*xfs_btree_bload_get_fn)(struct xfs_btree_cur *cur, void *priv);
+typedef int (*xfs_btree_bload_alloc_block_fn)(struct xfs_btree_cur *cur,
+		union xfs_btree_ptr *ptr, void *priv);
+typedef size_t (*xfs_btree_bload_iroot_size_fn)(struct xfs_btree_cur *cur,
+		unsigned int nr_this_level, void *priv);
+
+/* Bulk loading of staged btrees. */
+struct xfs_btree_bload {
+	/* Function to store a record in the cursor. */
+	xfs_btree_bload_get_fn		get_data;
+
+	/* Function to allocate a block for the btree. */
+	xfs_btree_bload_alloc_block_fn	alloc_block;
+
+	/* Function to compute the size of the in-core btree root block. */
+	xfs_btree_bload_iroot_size_fn	iroot_size;
+
+	/* Number of records the caller wants to store. */
+	uint64_t			nr_records;
+
+	/* Number of btree blocks needed to store those records. */
+	uint64_t			nr_blocks;
+
+	/*
+	 * Number of free records to leave in each leaf block.  If this (or
+	 * any of the slack values) are negative, this will be computed to
+	 * be halfway between maxrecs and minrecs.  This typically leaves the
+	 * block 75% full.
+	 */
+	int				leaf_slack;
+
+	/* Number of free keyptrs to leave in each node block. */
+	int				node_slack;
+
+	/* Computed btree height. */
+	unsigned int			btree_height;
+};
+
+int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
+		struct xfs_btree_bload *bbl, uint64_t nr_records);
+int xfs_btree_bload(struct xfs_btree_cur *cur, struct xfs_btree_bload *bbl,
+		void *priv);
+
 #endif	/* __XFS_BTREE_H__ */
diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
index bc85b89f88ca..9b5e58a92381 100644
--- a/fs/xfs/xfs_trace.c
+++ b/fs/xfs/xfs_trace.c
@@ -6,6 +6,7 @@ 
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_shared.h"
+#include "xfs_bit.h"
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index a78055521fcd..6d7ba64b7a0f 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -35,6 +35,7 @@  struct xfs_icreate_log;
 struct xfs_owner_info;
 struct xfs_trans_res;
 struct xfs_inobt_rec_incore;
+union xfs_btree_ptr;
 
 DECLARE_EVENT_CLASS(xfs_attr_list_class,
 	TP_PROTO(struct xfs_attr_list_context *ctx),
@@ -3670,6 +3671,90 @@  TRACE_EVENT(xfs_btree_commit_ifakeroot,
 		  __entry->blocks)
 )
 
+TRACE_EVENT(xfs_btree_bload_level_geometry,
+	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
+		 uint64_t nr_this_level, unsigned int nr_per_block,
+		 unsigned int desired_npb, uint64_t blocks,
+		 uint64_t blocks_with_extra),
+	TP_ARGS(cur, level, nr_this_level, nr_per_block, desired_npb, blocks,
+		blocks_with_extra),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_btnum_t, btnum)
+		__field(unsigned int, level)
+		__field(unsigned int, nlevels)
+		__field(uint64_t, nr_this_level)
+		__field(unsigned int, nr_per_block)
+		__field(unsigned int, desired_npb)
+		__field(unsigned long long, blocks)
+		__field(unsigned long long, blocks_with_extra)
+	),
+	TP_fast_assign(
+		__entry->dev = cur->bc_mp->m_super->s_dev;
+		__entry->btnum = cur->bc_btnum;
+		__entry->level = level;
+		__entry->nlevels = cur->bc_nlevels;
+		__entry->nr_this_level = nr_this_level;
+		__entry->nr_per_block = nr_per_block;
+		__entry->desired_npb = desired_npb;
+		__entry->blocks = blocks;
+		__entry->blocks_with_extra = blocks_with_extra;
+	),
+	TP_printk("dev %d:%d btree %s level %u/%u nr_this_level %llu nr_per_block %u desired_npb %u blocks %llu blocks_with_extra %llu",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
+		  __entry->level,
+		  __entry->nlevels,
+		  __entry->nr_this_level,
+		  __entry->nr_per_block,
+		  __entry->desired_npb,
+		  __entry->blocks,
+		  __entry->blocks_with_extra)
+)
+
+TRACE_EVENT(xfs_btree_bload_block,
+	TP_PROTO(struct xfs_btree_cur *cur, unsigned int level,
+		 uint64_t block_idx, uint64_t nr_blocks,
+		 union xfs_btree_ptr *ptr, unsigned int nr_records),
+	TP_ARGS(cur, level, block_idx, nr_blocks, ptr, nr_records),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_btnum_t, btnum)
+		__field(unsigned int, level)
+		__field(unsigned long long, block_idx)
+		__field(unsigned long long, nr_blocks)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agblock_t, agbno)
+		__field(unsigned int, nr_records)
+	),
+	TP_fast_assign(
+		__entry->dev = cur->bc_mp->m_super->s_dev;
+		__entry->btnum = cur->bc_btnum;
+		__entry->level = level;
+		__entry->block_idx = block_idx;
+		__entry->nr_blocks = nr_blocks;
+		if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
+			xfs_fsblock_t	fsb = be64_to_cpu(ptr->l);
+
+			__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsb);
+			__entry->agbno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsb);
+		} else {
+			__entry->agno = cur->bc_private.a.agno;
+			__entry->agbno = be32_to_cpu(ptr->s);
+		}
+		__entry->nr_records = nr_records;
+	),
+	TP_printk("dev %d:%d btree %s level %u block %llu/%llu fsb (%u/%u) recs %u",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __print_symbolic(__entry->btnum, XFS_BTNUM_STRINGS),
+		  __entry->level,
+		  __entry->block_idx,
+		  __entry->nr_blocks,
+		  __entry->agno,
+		  __entry->agbno,
+		  __entry->nr_records)
+)
+
 #endif /* _TRACE_XFS_H */
 
 #undef TRACE_INCLUDE_PATH