Merge branch 'for-linus' of git://oss.sgi.com/xfs/xfs

* 'for-linus' of git://oss.sgi.com/xfs/xfs:
  xfs: Ensure inode allocation buffers are fully replayed
  xfs: enable background pushing of the CIL
  xfs: forced unmounts need to push the CIL
  xfs: Introduce delayed logging core code
  xfs: Delayed logging design documentation
  xfs: Improve scalability of busy extent tracking
  xfs: make the log ticket ID available outside the log infrastructure
  xfs: clean up log ticket overrun debug output
  xfs: Clean up XFS_BLI_* flag namespace
  xfs: modify buffer item reference counting
  xfs: allow log ticket allocation to take allocation flags
  xfs: Don't reuse the same transaction ID for duplicated transactions.

Author: torvalds

Documentation/filesystems/xfs-delayed-logging-design.txt

@@ -77,6 +77,7 @@ xfs-y				+= xfs_alloc.o \
 				   xfs_itable.o \
 				   xfs_dfrag.o \
 				   xfs_log.o \
+				   xfs_log_cil.o \
 				   xfs_log_recover.o \
 				   xfs_mount.o \
 				   xfs_mru_cache.o \

fs/xfs/Makefile

@@ -37,6 +37,7 @@
 
 #include "xfs_sb.h"
 #include "xfs_inum.h"
+#include "xfs_log.h"
 #include "xfs_ag.h"
 #include "xfs_dmapi.h"
 #include "xfs_mount.h"
@@ -850,13 +851,21 @@ xfs_buf_lock_value(
  *	Note that this in no way locks the underlying pages, so it is only
  *	useful for synchronizing concurrent use of buffer objects, not for
  *	synchronizing independent access to the underlying pages.
+ *
+ *	If we come across a stale, pinned, locked buffer, we know that we
+ *	are being asked to lock a buffer that has been reallocated. Because
+ *	it is pinned, we know that the log has not been pushed to disk and
+ *	hence it will still be locked. Rather than sleeping until someone
+ *	else pushes the log, push it ourselves before trying to get the lock.
  */
 void
 xfs_buf_lock(
 	xfs_buf_t		*bp)
 {
 	trace_xfs_buf_lock(bp, _RET_IP_);
 
+	if (atomic_read(&bp->b_pin_count) && (bp->b_flags & XBF_STALE))
+		xfs_log_force(bp->b_mount, 0);
 	if (atomic_read(&bp->b_io_remaining))
 		blk_run_address_space(bp->b_target->bt_mapping);
 	down(&bp->b_sema);

fs/xfs/linux-2.6/xfs_buf.c

@@ -19,6 +19,7 @@
 #include "xfs_dmapi.h"
 #include "xfs_sb.h"
 #include "xfs_inum.h"
+#include "xfs_log.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
 #include "xfs_quota.h"

fs/xfs/linux-2.6/xfs_quotaops.c

@@ -119,6 +119,8 @@ mempool_t *xfs_ioend_pool;
 #define MNTOPT_DMAPI	"dmapi"		/* DMI enabled (DMAPI / XDSM) */
 #define MNTOPT_XDSM	"xdsm"		/* DMI enabled (DMAPI / XDSM) */
 #define MNTOPT_DMI	"dmi"		/* DMI enabled (DMAPI / XDSM) */
+#define MNTOPT_DELAYLOG   "delaylog"	/* Delayed loging enabled */
+#define MNTOPT_NODELAYLOG "nodelaylog"	/* Delayed loging disabled */
 
 /*
  * Table driven mount option parser.
@@ -374,6 +376,13 @@ xfs_parseargs(
 			mp->m_flags |= XFS_MOUNT_DMAPI;
 		} else if (!strcmp(this_char, MNTOPT_DMI)) {
 			mp->m_flags |= XFS_MOUNT_DMAPI;
+		} else if (!strcmp(this_char, MNTOPT_DELAYLOG)) {
+			mp->m_flags |= XFS_MOUNT_DELAYLOG;
+			cmn_err(CE_WARN,
+				"Enabling EXPERIMENTAL delayed logging feature "
+				"- use at your own risk.\n");
+		} else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) {
+			mp->m_flags &= ~XFS_MOUNT_DELAYLOG;
 		} else if (!strcmp(this_char, "ihashsize")) {
 			cmn_err(CE_WARN,
 	"XFS: ihashsize no longer used, option is deprecated.");
@@ -535,6 +544,7 @@ xfs_showargs(
 		{ XFS_MOUNT_FILESTREAMS,	"," MNTOPT_FILESTREAM },
 		{ XFS_MOUNT_DMAPI,		"," MNTOPT_DMAPI },
 		{ XFS_MOUNT_GRPID,		"," MNTOPT_GRPID },
+		{ XFS_MOUNT_DELAYLOG,		"," MNTOPT_DELAYLOG },
 		{ 0, NULL }
 	};
 	static struct proc_xfs_info xfs_info_unset[] = {
@@ -1755,7 +1765,7 @@ xfs_init_zones(void)
 	 * but it is much faster.
 	 */
 	xfs_buf_item_zone = kmem_zone_init((sizeof(xfs_buf_log_item_t) +
-				(((XFS_MAX_BLOCKSIZE / XFS_BLI_CHUNK) /
+				(((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) /
 				  NBWORD) * sizeof(int))), "xfs_buf_item");
 	if (!xfs_buf_item_zone)
 		goto out_destroy_trans_zone;

fs/xfs/linux-2.6/xfs_super.c

@@ -1059,83 +1059,112 @@ TRACE_EVENT(xfs_bunmap,
 
 );
 
+#define XFS_BUSY_SYNC \
+	{ 0,	"async" }, \
+	{ 1,	"sync" }
+
 TRACE_EVENT(xfs_alloc_busy,
-	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno,
-		 xfs_extlen_t len, int slot),
-	TP_ARGS(mp, agno, agbno, len, slot),
+	TP_PROTO(struct xfs_trans *trans, xfs_agnumber_t agno,
+		 xfs_agblock_t agbno, xfs_extlen_t len, int sync),
+	TP_ARGS(trans, agno, agbno, len, sync),
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
+		__field(struct xfs_trans *, tp)
+		__field(int, tid)
 		__field(xfs_agnumber_t, agno)
 		__field(xfs_agblock_t, agbno)
 		__field(xfs_extlen_t, len)
-		__field(int, slot)
+		__field(int, sync)
 	),
 	TP_fast_assign(
-		__entry->dev = mp->m_super->s_dev;
+		__entry->dev = trans->t_mountp->m_super->s_dev;
+		__entry->tp = trans;
+		__entry->tid = trans->t_ticket->t_tid;
 		__entry->agno = agno;
 		__entry->agbno = agbno;
 		__entry->len = len;
-		__entry->slot = slot;
+		__entry->sync = sync;
 	),
-	TP_printk("dev %d:%d agno %u agbno %u len %u slot %d",
+	TP_printk("dev %d:%d trans 0x%p tid 0x%x agno %u agbno %u len %u %s",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->tp,
+		  __entry->tid,
 		  __entry->agno,
 		  __entry->agbno,
 		  __entry->len,
-		  __entry->slot)
+		  __print_symbolic(__entry->sync, XFS_BUSY_SYNC))
 
 );
 
-#define XFS_BUSY_STATES \
-	{ 0,	"found" }, \
-	{ 1,	"missing" }
-
 TRACE_EVENT(xfs_alloc_unbusy,
 	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
-		 int slot, int found),
-	TP_ARGS(mp, agno, slot, found),
+		 xfs_agblock_t agbno, xfs_extlen_t len),
+	TP_ARGS(mp, agno, agbno, len),
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
 		__field(xfs_agnumber_t, agno)
-		__field(int, slot)
-		__field(int, found)
+		__field(xfs_agblock_t, agbno)
+		__field(xfs_extlen_t, len)
 	),
 	TP_fast_assign(
 		__entry->dev = mp->m_super->s_dev;
 		__entry->agno = agno;
-		__entry->slot = slot;
-		__entry->found = found;
+		__entry->agbno = agbno;
+		__entry->len = len;
 	),
-	TP_printk("dev %d:%d agno %u slot %d %s",
+	TP_printk("dev %d:%d agno %u agbno %u len %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
-		  __entry->slot,
-		  __print_symbolic(__entry->found, XFS_BUSY_STATES))
+		  __entry->agbno,
+		  __entry->len)
 );
 
+#define XFS_BUSY_STATES \
+	{ 0,	"missing" }, \
+	{ 1,	"found" }
+
 TRACE_EVENT(xfs_alloc_busysearch,
-	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno,
-		 xfs_extlen_t len, xfs_lsn_t lsn),
-	TP_ARGS(mp, agno, agbno, len, lsn),
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
+		 xfs_agblock_t agbno, xfs_extlen_t len, int found),
+	TP_ARGS(mp, agno, agbno, len, found),
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
 		__field(xfs_agnumber_t, agno)
 		__field(xfs_agblock_t, agbno)
 		__field(xfs_extlen_t, len)
-		__field(xfs_lsn_t, lsn)
+		__field(int, found)
 	),
 	TP_fast_assign(
 		__entry->dev = mp->m_super->s_dev;
 		__entry->agno = agno;
 		__entry->agbno = agbno;
 		__entry->len = len;
-		__entry->lsn = lsn;
+		__entry->found = found;
 	),
-	TP_printk("dev %d:%d agno %u agbno %u len %u force lsn 0x%llx",
+	TP_printk("dev %d:%d agno %u agbno %u len %u %s",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
 		  __entry->agbno,
 		  __entry->len,
+		  __print_symbolic(__entry->found, XFS_BUSY_STATES))
+);
+
+TRACE_EVENT(xfs_trans_commit_lsn,
+	TP_PROTO(struct xfs_trans *trans),
+	TP_ARGS(trans),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(struct xfs_trans *, tp)
+		__field(xfs_lsn_t, lsn)
+	),
+	TP_fast_assign(
+		__entry->dev = trans->t_mountp->m_super->s_dev;
+		__entry->tp = trans;
+		__entry->lsn = trans->t_commit_lsn;
+	),
+	TP_printk("dev %d:%d trans 0x%p commit_lsn 0x%llx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->tp,
 		  __entry->lsn)
 );
 

fs/xfs/linux-2.6/xfs_trace.h

@@ -344,9 +344,9 @@ xfs_qm_init_dquot_blk(
 	for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++)
 		xfs_qm_dqinit_core(curid, type, d);
 	xfs_trans_dquot_buf(tp, bp,
-			    (type & XFS_DQ_USER ? XFS_BLI_UDQUOT_BUF :
-			    ((type & XFS_DQ_PROJ) ? XFS_BLI_PDQUOT_BUF :
-			     XFS_BLI_GDQUOT_BUF)));
+			    (type & XFS_DQ_USER ? XFS_BLF_UDQUOT_BUF :
+			    ((type & XFS_DQ_PROJ) ? XFS_BLF_PDQUOT_BUF :
+			     XFS_BLF_GDQUOT_BUF)));
 	xfs_trans_log_buf(tp, bp, 0, BBTOB(q->qi_dqchunklen) - 1);
 }
 

fs/xfs/quota/xfs_dquot.c

@@ -175,14 +175,20 @@ typedef struct xfs_agfl {
 } xfs_agfl_t;
 
 /*
- * Busy block/extent entry.  Used in perag to mark blocks that have been freed
- * but whose transactions aren't committed to disk yet.
+ * Busy block/extent entry.  Indexed by a rbtree in perag to mark blocks that
+ * have been freed but whose transactions aren't committed to disk yet.
+ *
+ * Note that we use the transaction ID to record the transaction, not the
+ * transaction structure itself. See xfs_alloc_busy_insert() for details.
  */
-typedef struct xfs_perag_busy {
-	xfs_agblock_t	busy_start;
-	xfs_extlen_t	busy_length;
-	struct xfs_trans *busy_tp;	/* transaction that did the free */
-} xfs_perag_busy_t;
+struct xfs_busy_extent {
+	struct rb_node	rb_node;	/* ag by-bno indexed search tree */
+	struct list_head list;		/* transaction busy extent list */
+	xfs_agnumber_t	agno;
+	xfs_agblock_t	bno;
+	xfs_extlen_t	length;
+	xlog_tid_t	tid;		/* transaction that created this */
+};
 
 /*
  * Per-ag incore structure, copies of information in agf and agi,
@@ -216,7 +222,8 @@ typedef struct xfs_perag {
 	xfs_agino_t	pagl_leftrec;
 	xfs_agino_t	pagl_rightrec;
 #ifdef __KERNEL__
-	spinlock_t	pagb_lock;	/* lock for pagb_list */
+	spinlock_t	pagb_lock;	/* lock for pagb_tree */
+	struct rb_root	pagb_tree;	/* ordered tree of busy extents */
 
 	atomic_t        pagf_fstrms;    /* # of filestreams active in this AG */
 
@@ -226,7 +233,6 @@ typedef struct xfs_perag {
 	int		pag_ici_reclaimable;	/* reclaimable inodes */
 #endif
 	int		pagb_count;	/* pagb slots in use */
-	xfs_perag_busy_t pagb_list[XFS_PAGB_NUM_SLOTS];	/* unstable blocks */
 } xfs_perag_t;
 
 /*