All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: david@fromorbit.com, darrick.wong@oracle.com
Cc: linux-xfs@vger.kernel.org, xfs@oss.sgi.com
Subject: [PATCH 08/29] xfs: add realtime rmap btree operations
Date: Thu, 25 Aug 2016 16:59:26 -0700	[thread overview]
Message-ID: <147216956639.7022.7959437975156604082.stgit@birch.djwong.org> (raw)
In-Reply-To: <147216950911.7022.438115723996286926.stgit@birch.djwong.org>

Implement the generic btree operations needed to manipulate rtrmap
btree blocks. This is different from the regular rmapbt in that we
allocate space from the filesystem at large, and are neither
constrained to the free space nor any particular AG.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 libxfs/xfs_rtrmap_btree.c |  319 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 319 insertions(+)


diff --git a/libxfs/xfs_rtrmap_btree.c b/libxfs/xfs_rtrmap_btree.c
index c62e5c8..66c9190 100644
--- a/libxfs/xfs_rtrmap_btree.c
+++ b/libxfs/xfs_rtrmap_btree.c
@@ -31,10 +31,12 @@
 #include "xfs_trans.h"
 #include "xfs_alloc.h"
 #include "xfs_btree.h"
+#include "xfs_rmap.h"
 #include "xfs_rtrmap_btree.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
 #include "xfs_ag_resv.h"
+#include "xfs_bmap.h"
 
 /*
  * Realtime Reverse map btree.
@@ -68,6 +70,275 @@ xfs_rtrmapbt_dup_cursor(
 	return new;
 }
 
+STATIC int
+xfs_rtrmapbt_alloc_block(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_ptr	*start,
+	union xfs_btree_ptr	*new,
+	int			*stat)
+{
+	struct xfs_alloc_arg	args;
+	int			error;
+
+	memset(&args, 0, sizeof(args));
+	args.tp = cur->bc_tp;
+	args.mp = cur->bc_mp;
+	args.fsbno = cur->bc_private.b.firstblock;
+	args.firstblock = args.fsbno;
+	xfs_rmap_ino_bmbt_owner(&args.oinfo, cur->bc_private.b.ip->i_ino,
+			cur->bc_private.b.whichfork);
+
+	if (args.fsbno == NULLFSBLOCK) {
+		args.fsbno = be64_to_cpu(start->l);
+		args.type = XFS_ALLOCTYPE_START_BNO;
+		/*
+		 * Make sure there is sufficient room left in the AG to
+		 * complete a full tree split for an extent insert.  If
+		 * we are converting the middle part of an extent then
+		 * we may need space for two tree splits.
+		 *
+		 * We are relying on the caller to make the correct block
+		 * reservation for this operation to succeed.  If the
+		 * reservation amount is insufficient then we may fail a
+		 * block allocation here and corrupt the filesystem.
+		 */
+		args.minleft = args.tp->t_blk_res;
+	} else if (cur->bc_private.b.dfops->dop_low) {
+		args.type = XFS_ALLOCTYPE_START_BNO;
+	} else {
+		args.type = XFS_ALLOCTYPE_NEAR_BNO;
+	}
+
+	args.minlen = args.maxlen = args.prod = 1;
+	args.wasdel = 0;
+	error = xfs_alloc_vextent(&args);
+	if (error)
+		goto error0;
+
+	if (args.fsbno == NULLFSBLOCK && args.minleft) {
+		/*
+		 * Could not find an AG with enough free space to satisfy
+		 * a full btree split.  Try again without minleft and if
+		 * successful activate the lowspace algorithm.
+		 */
+		args.fsbno = 0;
+		args.type = XFS_ALLOCTYPE_FIRST_AG;
+		args.minleft = 0;
+		error = xfs_alloc_vextent(&args);
+		if (error)
+			goto error0;
+		cur->bc_private.b.dfops->dop_low = true;
+	}
+	if (args.fsbno == NULLFSBLOCK) {
+		XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT);
+		*stat = 0;
+		return 0;
+	}
+	ASSERT(args.len == 1);
+	cur->bc_private.b.firstblock = args.fsbno;
+	cur->bc_private.b.allocated++;
+	cur->bc_private.b.ip->i_d.di_nblocks++;
+	xfs_trans_log_inode(args.tp, cur->bc_private.b.ip, XFS_ILOG_CORE);
+
+	new->l = cpu_to_be64(args.fsbno);
+
+	XFS_BTREE_TRACE_CURSOR(cur, XBT_EXIT);
+	*stat = 1;
+	return 0;
+
+ error0:
+	XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR);
+	return error;
+}
+
+STATIC int
+xfs_rtrmapbt_free_block(
+	struct xfs_btree_cur	*cur,
+	struct xfs_buf		*bp)
+{
+	struct xfs_mount	*mp = cur->bc_mp;
+	struct xfs_inode	*ip = cur->bc_private.b.ip;
+	struct xfs_trans	*tp = cur->bc_tp;
+	xfs_fsblock_t		fsbno = XFS_DADDR_TO_FSB(mp, XFS_BUF_ADDR(bp));
+	struct xfs_owner_info	oinfo;
+
+	xfs_rmap_ino_bmbt_owner(&oinfo, ip->i_ino, cur->bc_private.b.whichfork);
+	xfs_bmap_add_free(mp, cur->bc_private.b.dfops, fsbno, 1, &oinfo);
+	ip->i_d.di_nblocks--;
+
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+	return 0;
+}
+
+/*
+ * Calculate number of records in the in-core realtime rmap btree inode root.
+ */
+STATIC int
+xfs_rtrmapbt_broot_maxrecs(
+	int			blocklen,
+	bool			leaf)
+{
+	blocklen -= XFS_RTRMAP_BLOCK_LEN;
+
+	if (leaf)
+		return blocklen / sizeof(struct xfs_rtrmap_rec);
+	return blocklen / (2 * sizeof(struct xfs_rtrmap_key) +
+			sizeof(xfs_rtrmap_ptr_t));
+}
+
+STATIC int
+xfs_rtrmapbt_get_minrecs(
+	struct xfs_btree_cur	*cur,
+	int			level)
+{
+	struct xfs_ifork	*ifp;
+
+	if (level == cur->bc_nlevels - 1) {
+		ifp = XFS_IFORK_PTR(cur->bc_private.b.ip,
+				    cur->bc_private.b.whichfork);
+
+		return xfs_rtrmapbt_broot_maxrecs(ifp->if_broot_bytes,
+				level == 0) / 2;
+	}
+
+	return cur->bc_mp->m_rtrmap_mnr[level != 0];
+}
+
+STATIC int
+xfs_rtrmapbt_get_maxrecs(
+	struct xfs_btree_cur	*cur,
+	int			level)
+{
+	struct xfs_ifork	*ifp;
+
+	if (level == cur->bc_nlevels - 1) {
+		ifp = XFS_IFORK_PTR(cur->bc_private.b.ip,
+				    cur->bc_private.b.whichfork);
+
+		return xfs_rtrmapbt_broot_maxrecs(ifp->if_broot_bytes,
+				level == 0);
+	}
+
+	return cur->bc_mp->m_rtrmap_mxr[level != 0];
+}
+
+STATIC void
+xfs_rtrmapbt_init_key_from_rec(
+	union xfs_btree_key	*key,
+	union xfs_btree_rec	*rec)
+{
+	key->rtrmap.rm_startblock = rec->rtrmap.rm_startblock;
+	key->rtrmap.rm_owner = rec->rtrmap.rm_owner;
+	key->rtrmap.rm_offset = rec->rtrmap.rm_offset;
+}
+
+STATIC void
+xfs_rtrmapbt_init_high_key_from_rec(
+	union xfs_btree_key	*key,
+	union xfs_btree_rec	*rec)
+{
+	__uint64_t		off;
+	int			adj;
+
+	adj = be64_to_cpu(rec->rtrmap.rm_blockcount) - 1;
+
+	key->rtrmap.rm_startblock = rec->rtrmap.rm_startblock;
+	be64_add_cpu(&key->rtrmap.rm_startblock, adj);
+	key->rtrmap.rm_owner = rec->rtrmap.rm_owner;
+	key->rtrmap.rm_offset = rec->rtrmap.rm_offset;
+	if (XFS_RMAP_NON_INODE_OWNER(be64_to_cpu(rec->rtrmap.rm_owner)) ||
+	    XFS_RMAP_IS_BMBT_BLOCK(be64_to_cpu(rec->rtrmap.rm_offset)))
+		return;
+	off = be64_to_cpu(key->rtrmap.rm_offset);
+	off = (XFS_RMAP_OFF(off) + adj) | (off & ~XFS_RMAP_OFF_MASK);
+	key->rtrmap.rm_offset = cpu_to_be64(off);
+}
+
+STATIC void
+xfs_rtrmapbt_init_rec_from_cur(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_rec	*rec)
+{
+	rec->rtrmap.rm_startblock = cpu_to_be64(cur->bc_rec.r.rm_startblock);
+	rec->rtrmap.rm_blockcount = cpu_to_be64(cur->bc_rec.r.rm_blockcount);
+	rec->rtrmap.rm_owner = cpu_to_be64(cur->bc_rec.r.rm_owner);
+	rec->rtrmap.rm_offset = cpu_to_be64(
+			xfs_rmap_irec_offset_pack(&cur->bc_rec.r));
+}
+
+STATIC void
+xfs_rtrmapbt_init_ptr_from_cur(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_ptr	*ptr)
+{
+	ptr->l = 0;
+}
+
+STATIC __int64_t
+xfs_rtrmapbt_key_diff(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_key	*key)
+{
+	struct xfs_rmap_irec	*rec = &cur->bc_rec.r;
+	struct xfs_rtrmap_key	*kp = &key->rtrmap;
+	__u64			x, y;
+
+	x = be64_to_cpu(kp->rm_startblock);
+	y = rec->rm_startblock;
+	if (x > y)
+		return 1;
+	else if (y > x)
+		return -1;
+
+	x = be64_to_cpu(kp->rm_owner);
+	y = rec->rm_owner;
+	if (x > y)
+		return 1;
+	else if (y > x)
+		return -1;
+
+	x = XFS_RMAP_OFF(be64_to_cpu(kp->rm_offset));
+	y = rec->rm_offset;
+	if (x > y)
+		return 1;
+	else if (y > x)
+		return -1;
+	return 0;
+}
+
+STATIC __int64_t
+xfs_rtrmapbt_diff_two_keys(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_key	*k1,
+	union xfs_btree_key	*k2)
+{
+	struct xfs_rtrmap_key	*kp1 = &k1->rtrmap;
+	struct xfs_rtrmap_key	*kp2 = &k2->rtrmap;
+	__u64			x, y;
+
+	x = be64_to_cpu(kp1->rm_startblock);
+	y = be64_to_cpu(kp2->rm_startblock);
+	if (x > y)
+		return 1;
+	else if (y > x)
+		return -1;
+
+	x = be64_to_cpu(kp1->rm_owner);
+	y = be64_to_cpu(kp2->rm_owner);
+	if (x > y)
+		return 1;
+	else if (y > x)
+		return -1;
+
+	x = XFS_RMAP_OFF(be64_to_cpu(kp1->rm_offset));
+	y = XFS_RMAP_OFF(be64_to_cpu(kp2->rm_offset));
+	if (x > y)
+		return 1;
+	else if (y > x)
+		return -1;
+	return 0;
+}
+
 static bool
 xfs_rtrmapbt_verify(
 	struct xfs_buf		*bp)
@@ -125,12 +396,60 @@ const struct xfs_buf_ops xfs_rtrmapbt_buf_ops = {
 	.verify_write		= xfs_rtrmapbt_write_verify,
 };
 
+STATIC int
+xfs_rtrmapbt_keys_inorder(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_key	*k1,
+	union xfs_btree_key	*k2)
+{
+	if (be64_to_cpu(k1->rtrmap.rm_startblock) <
+	    be64_to_cpu(k2->rtrmap.rm_startblock))
+		return 1;
+	if (be64_to_cpu(k1->rtrmap.rm_owner) <
+	    be64_to_cpu(k2->rtrmap.rm_owner))
+		return 1;
+	if (XFS_RMAP_OFF(be64_to_cpu(k1->rtrmap.rm_offset)) <=
+	    XFS_RMAP_OFF(be64_to_cpu(k2->rtrmap.rm_offset)))
+		return 1;
+	return 0;
+}
+
+STATIC int
+xfs_rtrmapbt_recs_inorder(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_rec	*r1,
+	union xfs_btree_rec	*r2)
+{
+	if (be64_to_cpu(r1->rtrmap.rm_startblock) <
+	    be64_to_cpu(r2->rtrmap.rm_startblock))
+		return 1;
+	if (XFS_RMAP_OFF(be64_to_cpu(r1->rtrmap.rm_offset)) <
+	    XFS_RMAP_OFF(be64_to_cpu(r2->rtrmap.rm_offset)))
+		return 1;
+	if (be64_to_cpu(r1->rtrmap.rm_owner) <=
+	    be64_to_cpu(r2->rtrmap.rm_owner))
+		return 1;
+	return 0;
+}
+
 static const struct xfs_btree_ops xfs_rtrmapbt_ops = {
 	.rec_len		= sizeof(struct xfs_rtrmap_rec),
 	.key_len		= 2 * sizeof(struct xfs_rtrmap_key),
 
 	.dup_cursor		= xfs_rtrmapbt_dup_cursor,
+	.alloc_block		= xfs_rtrmapbt_alloc_block,
+	.free_block		= xfs_rtrmapbt_free_block,
+	.get_minrecs		= xfs_rtrmapbt_get_minrecs,
+	.get_maxrecs		= xfs_rtrmapbt_get_maxrecs,
+	.init_key_from_rec	= xfs_rtrmapbt_init_key_from_rec,
+	.init_high_key_from_rec	= xfs_rtrmapbt_init_high_key_from_rec,
+	.init_rec_from_cur	= xfs_rtrmapbt_init_rec_from_cur,
+	.init_ptr_from_cur	= xfs_rtrmapbt_init_ptr_from_cur,
+	.key_diff		= xfs_rtrmapbt_key_diff,
 	.buf_ops		= &xfs_rtrmapbt_buf_ops,
+	.diff_two_keys		= xfs_rtrmapbt_diff_two_keys,
+	.keys_inorder		= xfs_rtrmapbt_keys_inorder,
+	.recs_inorder		= xfs_rtrmapbt_recs_inorder,
 };
 
 /*

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

  parent reply	other threads:[~2016-08-25 23:59 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-25 23:58 [PATCH v8 00/29] xfsprogs: add realtime reverse-mapping support Darrick J. Wong
2016-08-25 23:58 ` [PATCH 01/29] xfs: refactor long-format btree header verification routines Darrick J. Wong
2016-08-25 23:58 ` [PATCH 02/29] xfs: make iroot_realloc a btree function Darrick J. Wong
2016-08-25 23:58 ` [PATCH 03/29] xfs: support storing records in the inode core root Darrick J. Wong
2016-08-25 23:59 ` [PATCH 04/29] xfs: widen xfs_refcount_irec fields to handle realtime rmapbt Darrick J. Wong
2016-08-25 23:59 ` [PATCH 05/29] xfs: introduce realtime rmap btree definitions Darrick J. Wong
2016-08-25 23:59 ` [PATCH 06/29] xfs: define the on-disk realtime rmap btree format Darrick J. Wong
2016-08-25 23:59 ` [PATCH 07/29] xfs: realtime rmap btree transaction reservations Darrick J. Wong
2016-08-25 23:59 ` Darrick J. Wong [this message]
2016-08-25 23:59 ` [PATCH 09/29] xfs: prepare rmap functions to deal with rtrmapbt Darrick J. Wong
2016-08-25 23:59 ` [PATCH 10/29] xfs: add a realtime flag to the rmap update log redo items Darrick J. Wong
2016-08-25 23:59 ` [PATCH 11/29] xfs: add realtime reverse map inode to superblock Darrick J. Wong
2016-08-25 23:59 ` [PATCH 12/29] xfs: wire up a new inode fork type for the realtime rmap Darrick J. Wong
2016-08-25 23:59 ` [PATCH 13/29] xfs: don't assume a left rmap when allocating a new rmap Darrick J. Wong
2016-08-26  0:00 ` [PATCH 14/29] xfs: wire up rmap map and unmap to the realtime rmapbt Darrick J. Wong
2016-08-26  0:00 ` [PATCH 15/29] xfs: scrub " Darrick J. Wong
2016-08-26  0:00 ` [PATCH 16/29] xfs_db: display the realtime rmap btree contents Darrick J. Wong
2016-08-26  0:00 ` [PATCH 17/29] xfs_db: support rudimentary checks of the rtrmap btree Darrick J. Wong
2016-08-26  0:00 ` [PATCH 18/29] xfs_db: copy the realtime rmap btree Darrick J. Wong
2016-08-26  0:00 ` [PATCH 19/29] xfs_db: make fsmap query the realtime reverse mapping tree Darrick J. Wong
2016-08-26  0:00 ` [PATCH 20/29] xfs_io: scrub the realtime rmap Darrick J. Wong
2016-08-26  0:00 ` [PATCH 21/29] xfs_repair: factor rrmapino into superblock size calculations Darrick J. Wong
2016-08-26  0:00 ` [PATCH 22/29] xfs_repair: pass private data pointer to scan_lbtree Darrick J. Wong
2016-08-26  0:01 ` [PATCH 23/29] xfs_repair: use realtime rmap btree data to check block types Darrick J. Wong
2016-08-26  0:01 ` [PATCH 24/29] xfs_repair: extend ag_rmap[] for realtime mapping collection Darrick J. Wong
2016-08-26  0:01 ` [PATCH 25/29] xfs_repair: collect relatime reverse-mapping data for refcount/rmap tree rebuilding Darrick J. Wong
2016-08-26  0:01 ` [PATCH 26/29] xfs_repair: check existing realtime rmapbt entries against observed rmaps Darrick J. Wong
2016-08-26  0:01 ` [PATCH 27/29] xfs_repair: find and mark the rtrmapbt inode Darrick J. Wong
2016-08-26  0:01 ` [PATCH 28/29] xfs_repair: rebuild the realtime rmap btree Darrick J. Wong
2016-08-26  0:01 ` [PATCH 29/29] mkfs: create the realtime rmap inode Darrick J. Wong

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=147216956639.7022.7959437975156604082.stgit@birch.djwong.org \
    --to=darrick.wong@oracle.com \
    --cc=david@fromorbit.com \
    --cc=linux-xfs@vger.kernel.org \
    --cc=xfs@oss.sgi.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.