Update core ZFS code from build 121 to build 141.

1 files changed, 117 insertions, 66 deletions

diff --git a/module/zfs/dmu_traverse.c b/module/zfs/dmu_traverse.c
index 89cbfad29..429c76ae1 100644
--- a/module/zfs/dmu_traverse.c
+++ b/module/zfs/dmu_traverse.c
@@ -19,8 +19,7 @@
  * CDDL HEADER END
  */
 /*
- * Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
- * Use is subject to license terms.
+ * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
  */
 
 #include <sys/zfs_context.h>
@@ -33,16 +32,10 @@
 #include <sys/spa.h>
 #include <sys/zio.h>
 #include <sys/dmu_impl.h>
+#include <sys/sa.h>
+#include <sys/sa_impl.h>
 #include <sys/callb.h>
 
-#define	SET_BOOKMARK(zb, objset, object, level, blkid)  \
-{                                                       \
-	(zb)->zb_objset = objset;                       \
-	(zb)->zb_object = object;                       \
-	(zb)->zb_level = level;                         \
-	(zb)->zb_blkid = blkid;                         \
-}
-
 struct prefetch_data {
 	kmutex_t pd_mtx;
 	kcondvar_t pd_cv;
@@ -68,27 +61,28 @@ static int traverse_dnode(struct traverse_data *td, const dnode_phys_t *dnp,
     arc_buf_t *buf, uint64_t objset, uint64_t object);
 
 /* ARGSUSED */
-static void
+static int
 traverse_zil_block(zilog_t *zilog, blkptr_t *bp, void *arg, uint64_t claim_txg)
 {
 	struct traverse_data *td = arg;
 	zbookmark_t zb;
 
 	if (bp->blk_birth == 0)
-		return;
+		return (0);
 
 	if (claim_txg == 0 && bp->blk_birth >= spa_first_txg(td->td_spa))
-		return;
+		return (0);
+
+	SET_BOOKMARK(&zb, td->td_objset, ZB_ZIL_OBJECT, ZB_ZIL_LEVEL,
+	    bp->blk_cksum.zc_word[ZIL_ZC_SEQ]);
+
+	(void) td->td_func(td->td_spa, zilog, bp, NULL, &zb, NULL, td->td_arg);
 
-	zb.zb_objset = td->td_objset;
-	zb.zb_object = 0;
-	zb.zb_level = -1;
-	zb.zb_blkid = bp->blk_cksum.zc_word[ZIL_ZC_SEQ];
-	VERIFY(0 == td->td_func(td->td_spa, bp, &zb, NULL, td->td_arg));
+	return (0);
 }
 
 /* ARGSUSED */
-static void
+static int
 traverse_zil_record(zilog_t *zilog, lr_t *lrc, void *arg, uint64_t claim_txg)
 {
 	struct traverse_data *td = arg;
@@ -99,17 +93,18 @@ traverse_zil_record(zilog_t *zilog, lr_t *lrc, void *arg, uint64_t claim_txg)
 		zbookmark_t zb;
 
 		if (bp->blk_birth == 0)
-			return;
+			return (0);
 
 		if (claim_txg == 0 || bp->blk_birth < claim_txg)
-			return;
+			return (0);
 
-		zb.zb_objset = td->td_objset;
-		zb.zb_object = lr->lr_foid;
-		zb.zb_level = BP_GET_LEVEL(bp);
-		zb.zb_blkid = lr->lr_offset / BP_GET_LSIZE(bp);
-		VERIFY(0 == td->td_func(td->td_spa, bp, &zb, NULL, td->td_arg));
+		SET_BOOKMARK(&zb, td->td_objset, lr->lr_foid, ZB_ZIL_LEVEL,
+		    lr->lr_offset / BP_GET_LSIZE(bp));
+
+		(void) td->td_func(td->td_spa, zilog, bp, NULL, &zb, NULL,
+		    td->td_arg);
 	}
+	return (0);
 }
 
 static void
@@ -120,7 +115,7 @@ traverse_zil(struct traverse_data *td, zil_header_t *zh)
 
 	/*
 	 * We only want to visit blocks that have been claimed but not yet
-	 * replayed (or, in read-only mode, blocks that *would* be claimed).
+	 * replayed; plus, in read-only mode, blocks that are already stable.
 	 */
 	if (claim_txg == 0 && spa_writeable(td->td_spa))
 		return;
@@ -138,12 +133,14 @@ traverse_visitbp(struct traverse_data *td, const dnode_phys_t *dnp,
     arc_buf_t *pbuf, blkptr_t *bp, const zbookmark_t *zb)
 {
 	zbookmark_t czb;
-	int err = 0;
+	int err = 0, lasterr = 0;
 	arc_buf_t *buf = NULL;
 	struct prefetch_data *pd = td->td_pfd;
+	boolean_t hard = td->td_flags & TRAVERSE_HARD;
 
 	if (bp->blk_birth == 0) {
-		err = td->td_func(td->td_spa, NULL, zb, dnp, td->td_arg);
+		err = td->td_func(td->td_spa, NULL, NULL, pbuf, zb, dnp,
+		    td->td_arg);
 		return (err);
 	}
 
@@ -163,7 +160,8 @@ traverse_visitbp(struct traverse_data *td, const dnode_phys_t *dnp,
 	}
 
 	if (td->td_flags & TRAVERSE_PRE) {
-		err = td->td_func(td->td_spa, bp, zb, dnp, td->td_arg);
+		err = td->td_func(td->td_spa, NULL, bp, pbuf, zb, dnp,
+		    td->td_arg);
 		if (err)
 			return (err);
 	}
@@ -174,7 +172,7 @@ traverse_visitbp(struct traverse_data *td, const dnode_phys_t *dnp,
 		blkptr_t *cbp;
 		int epb = BP_GET_LSIZE(bp) >> SPA_BLKPTRSHIFT;
 
-		err = arc_read(NULL, td->td_spa, bp, pbuf,
+		err = dsl_read(NULL, td->td_spa, bp, pbuf,
 		    arc_getbuf_func, &buf,
 		    ZIO_PRIORITY_ASYNC_READ, ZIO_FLAG_CANFAIL, &flags, zb);
 		if (err)
@@ -187,15 +185,18 @@ traverse_visitbp(struct traverse_data *td, const dnode_phys_t *dnp,
 			    zb->zb_level - 1,
 			    zb->zb_blkid * epb + i);
 			err = traverse_visitbp(td, dnp, buf, cbp, &czb);
-			if (err)
-				break;
+			if (err) {
+				if (!hard)
+					break;
+				lasterr = err;
+			}
 		}
 	} else if (BP_GET_TYPE(bp) == DMU_OT_DNODE) {
 		uint32_t flags = ARC_WAIT;
 		int i;
 		int epb = BP_GET_LSIZE(bp) >> DNODE_SHIFT;
 
-		err = arc_read(NULL, td->td_spa, bp, pbuf,
+		err = dsl_read(NULL, td->td_spa, bp, pbuf,
 		    arc_getbuf_func, &buf,
 		    ZIO_PRIORITY_ASYNC_READ, ZIO_FLAG_CANFAIL, &flags, zb);
 		if (err)
@@ -203,18 +204,21 @@ traverse_visitbp(struct traverse_data *td, const dnode_phys_t *dnp,
 
 		/* recursively visitbp() blocks below this */
 		dnp = buf->b_data;
-		for (i = 0; i < epb && err == 0; i++, dnp++) {
+		for (i = 0; i < epb; i++, dnp++) {
 			err = traverse_dnode(td, dnp, buf, zb->zb_objset,
 			    zb->zb_blkid * epb + i);
-			if (err)
-				break;
+			if (err) {
+				if (!hard)
+					break;
+				lasterr = err;
+			}
 		}
 	} else if (BP_GET_TYPE(bp) == DMU_OT_OBJSET) {
 		uint32_t flags = ARC_WAIT;
 		objset_phys_t *osp;
 		dnode_phys_t *dnp;
 
-		err = arc_read_nolock(NULL, td->td_spa, bp,
+		err = dsl_read_nolock(NULL, td->td_spa, bp,
 		    arc_getbuf_func, &buf,
 		    ZIO_PRIORITY_ASYNC_READ, ZIO_FLAG_CANFAIL, &flags, zb);
 		if (err)
@@ -224,12 +228,21 @@ traverse_visitbp(struct traverse_data *td, const dnode_phys_t *dnp,
 		traverse_zil(td, &osp->os_zil_header);
 
 		dnp = &osp->os_meta_dnode;
-		err = traverse_dnode(td, dnp, buf, zb->zb_objset, 0);
+		err = traverse_dnode(td, dnp, buf, zb->zb_objset,
+		    DMU_META_DNODE_OBJECT);
+		if (err && hard) {
+			lasterr = err;
+			err = 0;
+		}
 		if (err == 0 && arc_buf_size(buf) >= sizeof (objset_phys_t)) {
 			dnp = &osp->os_userused_dnode;
 			err = traverse_dnode(td, dnp, buf, zb->zb_objset,
 			    DMU_USERUSED_OBJECT);
 		}
+		if (err && hard) {
+			lasterr = err;
+			err = 0;
+		}
 		if (err == 0 && arc_buf_size(buf) >= sizeof (objset_phys_t)) {
 			dnp = &osp->os_groupused_dnode;
 			err = traverse_dnode(td, dnp, buf, zb->zb_objset,
@@ -240,33 +253,52 @@ traverse_visitbp(struct traverse_data *td, const dnode_phys_t *dnp,
 	if (buf)
 		(void) arc_buf_remove_ref(buf, &buf);
 
-	if (err == 0 && (td->td_flags & TRAVERSE_POST))
-		err = td->td_func(td->td_spa, bp, zb, dnp, td->td_arg);
+	if (err == 0 && lasterr == 0 && (td->td_flags & TRAVERSE_POST)) {
+		err = td->td_func(td->td_spa, NULL, bp, pbuf, zb, dnp,
+		    td->td_arg);
+	}
 
-	return (err);
+	return (err != 0 ? err : lasterr);
 }
 
 static int
 traverse_dnode(struct traverse_data *td, const dnode_phys_t *dnp,
     arc_buf_t *buf, uint64_t objset, uint64_t object)
 {
-	int j, err = 0;
+	int j, err = 0, lasterr = 0;
 	zbookmark_t czb;
+	boolean_t hard = (td->td_flags & TRAVERSE_HARD);
 
 	for (j = 0; j < dnp->dn_nblkptr; j++) {
 		SET_BOOKMARK(&czb, objset, object, dnp->dn_nlevels - 1, j);
 		err = traverse_visitbp(td, dnp, buf,
 		    (blkptr_t *)&dnp->dn_blkptr[j], &czb);
-		if (err)
-			break;
+		if (err) {
+			if (!hard)
+				break;
+			lasterr = err;
+		}
 	}
-	return (err);
+
+	if (dnp->dn_flags & DNODE_FLAG_SPILL_BLKPTR) {
+		SET_BOOKMARK(&czb, objset,
+		    object, 0, DMU_SPILL_BLKID);
+		err = traverse_visitbp(td, dnp, buf,
+		    (blkptr_t *)&dnp->dn_spill, &czb);
+		if (err) {
+			if (!hard)
+				return (err);
+			lasterr = err;
+		}
+	}
+	return (err != 0 ? err : lasterr);
 }
 
 /* ARGSUSED */
 static int
-traverse_prefetcher(spa_t *spa, blkptr_t *bp, const zbookmark_t *zb,
-    const dnode_phys_t *dnp, void *arg)
+traverse_prefetcher(spa_t *spa, zilog_t *zilog, const blkptr_t *bp,
+    arc_buf_t *pbuf, const zbookmark_t *zb, const dnode_phys_t *dnp,
+    void *arg)
 {
 	struct prefetch_data *pfd = arg;
 	uint32_t aflags = ARC_NOWAIT | ARC_PREFETCH;
@@ -276,7 +308,8 @@ traverse_prefetcher(spa_t *spa, blkptr_t *bp, const zbookmark_t *zb,
 		return (EINTR);
 
 	if (bp == NULL || !((pfd->pd_flags & TRAVERSE_PREFETCH_DATA) ||
-	    BP_GET_TYPE(bp) == DMU_OT_DNODE || BP_GET_LEVEL(bp) > 0))
+	    BP_GET_TYPE(bp) == DMU_OT_DNODE || BP_GET_LEVEL(bp) > 0) ||
+	    BP_GET_TYPE(bp) == DMU_OT_INTENT_LOG)
 		return (0);
 
 	mutex_enter(&pfd->pd_mtx);
@@ -286,7 +319,7 @@ traverse_prefetcher(spa_t *spa, blkptr_t *bp, const zbookmark_t *zb,
 	cv_broadcast(&pfd->pd_cv);
 	mutex_exit(&pfd->pd_mtx);
 
-	(void) arc_read_nolock(NULL, spa, bp, NULL, NULL,
+	(void) dsl_read(NULL, spa, bp, pbuf, NULL, NULL,
 	    ZIO_PRIORITY_ASYNC_READ,
 	    ZIO_FLAG_CANFAIL | ZIO_FLAG_SPECULATIVE,
 	    &aflags, zb);
@@ -305,7 +338,8 @@ traverse_prefetch_thread(void *arg)
 	td.td_arg = td_main->td_pfd;
 	td.td_pfd = NULL;
 
-	SET_BOOKMARK(&czb, td.td_objset, 0, -1, 0);
+	SET_BOOKMARK(&czb, td.td_objset,
+	    ZB_ROOT_OBJECT, ZB_ROOT_LEVEL, ZB_ROOT_BLKID);
 	(void) traverse_visitbp(&td, NULL, NULL, td.td_rootbp, &czb);
 
 	mutex_enter(&td_main->td_pfd->pd_mtx);
@@ -346,7 +380,8 @@ traverse_impl(spa_t *spa, uint64_t objset, blkptr_t *rootbp,
 	    &td, TQ_NOQUEUE))
 		pd.pd_exited = B_TRUE;
 
-	SET_BOOKMARK(&czb, objset, 0, -1, 0);
+	SET_BOOKMARK(&czb, objset,
+	    ZB_ROOT_OBJECT, ZB_ROOT_LEVEL, ZB_ROOT_BLKID);
 	err = traverse_visitbp(&td, NULL, NULL, rootbp, &czb);
 
 	mutex_enter(&pd.pd_mtx);
@@ -378,43 +413,59 @@ traverse_dataset(dsl_dataset_t *ds, uint64_t txg_start, int flags,
  * NB: pool must not be changing on-disk (eg, from zdb or sync context).
  */
 int
-traverse_pool(spa_t *spa, blkptr_cb_t func, void *arg)
+traverse_pool(spa_t *spa, uint64_t txg_start, int flags,
+    blkptr_cb_t func, void *arg)
 {
-	int err;
+	int err, lasterr = 0;
 	uint64_t obj;
 	dsl_pool_t *dp = spa_get_dsl(spa);
 	objset_t *mos = dp->dp_meta_objset;
+	boolean_t hard = (flags & TRAVERSE_HARD);
 
 	/* visit the MOS */
 	err = traverse_impl(spa, 0, spa_get_rootblkptr(spa),
-	    0, TRAVERSE_PRE, func, arg);
+	    txg_start, flags, func, arg);
 	if (err)
 		return (err);
 
 	/* visit each dataset */
-	for (obj = 1; err == 0; err = dmu_object_next(mos, &obj, FALSE, 0)) {
+	for (obj = 1; err == 0 || (err != ESRCH && hard);
+	    err = dmu_object_next(mos, &obj, FALSE, txg_start)) {
 		dmu_object_info_t doi;
 
 		err = dmu_object_info(mos, obj, &doi);
-		if (err)
-			return (err);
+		if (err) {
+			if (!hard)
+				return (err);
+			lasterr = err;
+			continue;
+		}
 
 		if (doi.doi_type == DMU_OT_DSL_DATASET) {
 			dsl_dataset_t *ds;
+			uint64_t txg = txg_start;
+
 			rw_enter(&dp->dp_config_rwlock, RW_READER);
 			err = dsl_dataset_hold_obj(dp, obj, FTAG, &ds);
 			rw_exit(&dp->dp_config_rwlock);
-			if (err)
-				return (err);
-			err = traverse_dataset(ds,
-			    ds->ds_phys->ds_prev_snap_txg, TRAVERSE_PRE,
-			    func, arg);
+			if (err) {
+				if (!hard)
+					return (err);
+				lasterr = err;
+				continue;
+			}
+			if (ds->ds_phys->ds_prev_snap_txg > txg)
+				txg = ds->ds_phys->ds_prev_snap_txg;
+			err = traverse_dataset(ds, txg, flags, func, arg);
 			dsl_dataset_rele(ds, FTAG);
-			if (err)
-				return (err);
+			if (err) {
+				if (!hard)
+					return (err);
+				lasterr = err;
+			}
 		}
 	}
 	if (err == ESRCH)
 		err = 0;
-	return (err);
+	return (err != 0 ? err : lasterr);
 }