Illumos 5960, 5925

5960 zfs recv should prefetch indirect blocks
5925 zfs receive -o origin=
Reviewed by: Prakash Surya <prakash.surya@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>

References:
  https://www.illumos.org/issues/5960
  https://www.illumos.org/issues/5925
  illumos/illumos-gate@a2cdcdd

Porting notes:
- [lib/libzfs/libzfs_sendrecv.c]
  - b8864a2 Fix gcc cast warnings
  - 325f023 Add linux kernel device support
  - 5c3f61e Increase Linux pipe buffer size on 'zfs receive'
- [module/zfs/zfs_vnops.c]
  - 3558fd7 Prototype/structure update for Linux
  - c12e3a5 Restructure zfs_readdir() to fix regressions
- [module/zfs/zvol.c]
  - Function @zvol_map_block() isn't needed in ZoL
  - 9965059 Prefetch start and end of volumes
- [module/zfs/dmu.c]
  - Fixed ISO C90 - mixed declarations and code
  - Function dmu_prefetch() 'int i' is initialized before
    the following code block (c90 vs. c99)
- [module/zfs/dbuf.c]
  - fc5bb51 Fix stack dbuf_hold_impl()
  - 9b67f60 Illumos 4757, 4913
  - 34229a2 Reduce stack usage for recursive traverse_visitbp()
- [module/zfs/dmu_send.c]
  - Fixed ISO C90 - mixed declarations and code
  - b58986e Use large stacks when available
  - 241b541 Illumos 5959 - clean up per-dataset feature count code
  - 77aef6f Use vmem_alloc() for nvlists
  - 00b4602 Add linux kernel memory support

Ported-by: kernelOfTruth kerneloftruth@gmail.com
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>

Author: pcd1193182

cmd/zdb/zdb.c

@@ -2489,6 +2489,9 @@ zdb_blkptr_cb(spa_t *spa, zilog_t *zilog, const blkptr_t *bp,
 	dmu_object_type_t type;
 	boolean_t is_metadata;
 
+	if (bp == NULL)
+		return (0);
+
 	if (dump_opt['b'] >= 5 && bp->blk_birth > 0) {
 		char blkbuf[BP_SPRINTF_LEN];
 		snprintf_blkptr(blkbuf, sizeof (blkbuf), bp);
@@ -2985,7 +2988,7 @@ zdb_ddt_add_cb(spa_t *spa, zilog_t *zilog, const blkptr_t *bp,
 	avl_index_t where;
 	zdb_ddt_entry_t *zdde, zdde_search;
 
-	if (BP_IS_HOLE(bp) || BP_IS_EMBEDDED(bp))
+	if (bp == NULL || BP_IS_HOLE(bp) || BP_IS_EMBEDDED(bp))
 		return (0);
 
 	if (dump_opt['S'] > 1 && zb->zb_level == ZB_ROOT_LEVEL) {

cmd/zfs/zfs_main.c

@@ -249,8 +249,9 @@ get_usage(zfs_help_t idx)
 		return (gettext("\tpromote <clone-filesystem>\n"));
 	case HELP_RECEIVE:
 		return (gettext("\treceive [-vnFu] <filesystem|volume|"
-		"snapshot>\n"
-		"\treceive [-vnFu] [-d | -e] <filesystem>\n"));
+		    "snapshot>\n"
+		    "\treceive [-vnFu] [-o origin=<snapshot>] [-d | -e] "
+		    "<filesystem>\n"));
 	case HELP_RENAME:
 		return (gettext("\trename [-f] <filesystem|volume|snapshot> "
 		    "<filesystem|volume|snapshot>\n"
@@ -793,7 +794,7 @@ zfs_do_create(int argc, char **argv)
 				nomem();
 			break;
 		case 'o':
-			if (parseprop(props, optarg))
+			if (parseprop(props, optarg) != 0)
 				goto error;
 			break;
 		case 's':
@@ -3622,7 +3623,7 @@ zfs_do_snapshot(int argc, char **argv)
 	while ((c = getopt(argc, argv, "ro:")) != -1) {
 		switch (c) {
 		case 'o':
-			if (parseprop(props, optarg))
+			if (parseprop(props, optarg) != 0)
 				return (1);
 			break;
 		case 'r':
@@ -3881,10 +3882,19 @@ zfs_do_receive(int argc, char **argv)
 {
 	int c, err;
 	recvflags_t flags = { 0 };
+	nvlist_t *props;
+	nvpair_t *nvp = NULL;
+
+	if (nvlist_alloc(&props, NV_UNIQUE_NAME, 0) != 0)
+		nomem();
 
 	/* check options */
-	while ((c = getopt(argc, argv, ":denuvF")) != -1) {
+	while ((c = getopt(argc, argv, ":o:denuvF")) != -1) {
 		switch (c) {
+		case 'o':
+			if (parseprop(props, optarg) != 0)
+				return (1);
+			break;
 		case 'd':
 			flags.isprefix = B_TRUE;
 			break;
@@ -3929,6 +3939,13 @@ zfs_do_receive(int argc, char **argv)
 		usage(B_FALSE);
 	}
 
+	while ((nvp = nvlist_next_nvpair(props, nvp))) {
+		if (strcmp(nvpair_name(nvp), "origin") != 0) {
+			(void) fprintf(stderr, gettext("invalid option"));
+			usage(B_FALSE);
+		}
+	}
+
 	if (isatty(STDIN_FILENO)) {
 		(void) fprintf(stderr,
 		    gettext("Error: Backup stream can not be read "
@@ -3937,7 +3954,7 @@ zfs_do_receive(int argc, char **argv)
 		return (1);
 	}
 
-	err = zfs_receive(g_zfs, argv[0], &flags, STDIN_FILENO, NULL);
+	err = zfs_receive(g_zfs, argv[0], props, &flags, STDIN_FILENO, NULL);
 
 	return (err != 0);
 }

cmd/ztest/ztest.c

@@ -3728,7 +3728,8 @@ ztest_dmu_read_write(ztest_ds_t *zd, uint64_t id)
 	 */
 	n = ztest_random(regions) * stride + ztest_random(width);
 	s = 1 + ztest_random(2 * width - 1);
-	dmu_prefetch(os, bigobj, n * chunksize, s * chunksize);
+	dmu_prefetch(os, bigobj, 0, n * chunksize, s * chunksize,
+	    ZIO_PRIORITY_SYNC_READ);
 
 	/*
 	 * Pick a random index and compute the offsets into packobj and bigobj.
@@ -5930,8 +5931,10 @@ ztest_run(ztest_shared_t *zs)
 	 * Right before closing the pool, kick off a bunch of async I/O;
 	 * spa_close() should wait for it to complete.
 	 */
-	for (object = 1; object < 50; object++)
-		dmu_prefetch(spa->spa_meta_objset, object, 0, 1ULL << 20);
+	for (object = 1; object < 50; object++) {
+		dmu_prefetch(spa->spa_meta_objset, object, 0, 0, 1ULL << 20,
+		    ZIO_PRIORITY_SYNC_READ);
+	}
 
 	/* Verify that at least one commit cb was called in a timely fashion */
 	if (zc_cb_counter >= ZTEST_COMMIT_CB_MIN_REG)

include/libzfs.h

@@ -678,8 +678,8 @@ typedef struct recvflags {
 	boolean_t nomount;
 } recvflags_t;
 
-extern int zfs_receive(libzfs_handle_t *, const char *, recvflags_t *,
-    int, avl_tree_t *);
+extern int zfs_receive(libzfs_handle_t *, const char *, nvlist_t *,
+    recvflags_t *, int, avl_tree_t *);
 
 typedef enum diff_flags {
 	ZFS_DIFF_PARSEABLE = 0x1,

include/sys/Makefile.am

@@ -9,6 +9,7 @@ COMMON_H = \
 	$(top_srcdir)/include/sys/bplist.h \
 	$(top_srcdir)/include/sys/bpobj.h \
 	$(top_srcdir)/include/sys/bptree.h \
+	$(top_srcdir)/include/sys/bqueue.h \
 	$(top_srcdir)/include/sys/dbuf.h \
 	$(top_srcdir)/include/sys/ddt.h \
 	$(top_srcdir)/include/sys/dmu.h \
@@ -96,6 +97,7 @@ COMMON_H = \
 	$(top_srcdir)/include/sys/zio_compress.h \
 	$(top_srcdir)/include/sys/zio.h \
 	$(top_srcdir)/include/sys/zio_impl.h \
+	$(top_srcdir)/include/sys/zio_priority.h \
 	$(top_srcdir)/include/sys/zrlock.h
 
 KERNEL_H = \

include/sys/bqueue.h

@@ -0,0 +1,54 @@
+/*
+ * CDDL HEADER START
+ *
+ * This file and its contents are supplied under the terms of the
+ * Common Development and Distribution License ("CDDL"), version 1.0.
+ * You may only use this file in accordance with the terms of version
+ * 1.0 of the CDDL.
+ *
+ * A full copy of the text of the CDDL should have accompanied this
+ * source.  A copy of the CDDL is also available via the Internet at
+ * http://www.illumos.org/license/CDDL.
+ *
+ * CDDL HEADER END
+ */
+/*
+ * Copyright (c) 2014 by Delphix. All rights reserved.
+ */
+
+#ifndef	_BQUEUE_H
+#define	_BQUEUE_H
+
+#ifdef	__cplusplus
+extern "C" {
+#endif
+
+#include	<sys/zfs_context.h>
+
+typedef struct bqueue {
+	list_t bq_list;
+	kmutex_t bq_lock;
+	kcondvar_t bq_add_cv;
+	kcondvar_t bq_pop_cv;
+	uint64_t bq_size;
+	uint64_t bq_maxsize;
+	size_t bq_node_offset;
+} bqueue_t;
+
+typedef struct bqueue_node {
+	list_node_t bqn_node;
+	uint64_t bqn_size;
+} bqueue_node_t;
+
+
+int bqueue_init(bqueue_t *, uint64_t, size_t);
+void bqueue_destroy(bqueue_t *);
+void bqueue_enqueue(bqueue_t *, void *, uint64_t);
+void *bqueue_dequeue(bqueue_t *);
+boolean_t bqueue_empty(bqueue_t *);
+
+#ifdef	__cplusplus
+}
+#endif
+
+#endif	/* _BQUEUE_H */

include/sys/dbuf.h

@@ -261,8 +261,7 @@ typedef struct dbuf_hash_table {
 	kmutex_t hash_mutexes[DBUF_MUTEXES];
 } dbuf_hash_table_t;
 
-
-uint64_t dbuf_whichblock(struct dnode *di, uint64_t offset);
+uint64_t dbuf_whichblock(struct dnode *di, int64_t level, uint64_t offset);
 
 void dbuf_create_bonus(struct dnode *dn);
 int dbuf_spill_set_blksz(dmu_buf_t *db, uint64_t blksz, dmu_tx_t *tx);
@@ -272,10 +271,12 @@ void dbuf_rm_spill(struct dnode *dn, dmu_tx_t *tx);
 dmu_buf_impl_t *dbuf_hold(struct dnode *dn, uint64_t blkid, void *tag);
 dmu_buf_impl_t *dbuf_hold_level(struct dnode *dn, int level, uint64_t blkid,
     void *tag);
-int dbuf_hold_impl(struct dnode *dn, uint8_t level, uint64_t blkid, int create,
+int dbuf_hold_impl(struct dnode *dn, uint8_t level, uint64_t blkid,
+    boolean_t fail_sparse, boolean_t fail_uncached,
     void *tag, dmu_buf_impl_t **dbp);
 
-void dbuf_prefetch(struct dnode *dn, uint64_t blkid, zio_priority_t prio);
+void dbuf_prefetch(struct dnode *dn, int64_t level, uint64_t blkid,
+    zio_priority_t prio, arc_flags_t aflags);
 
 void dbuf_add_ref(dmu_buf_impl_t *db, void *tag);
 boolean_t dbuf_try_add_ref(dmu_buf_t *db, objset_t *os, uint64_t obj,

include/sys/dmu.h

@@ -44,6 +44,7 @@
 #include <sys/inttypes.h>
 #include <sys/cred.h>
 #include <sys/fs/zfs.h>
+#include <sys/zio_priority.h>
 #include <sys/uio.h>
 
 #ifdef	__cplusplus
@@ -737,8 +738,8 @@ extern int zfs_max_recordsize;
 /*
  * Asynchronously try to read in the data.
  */
-void dmu_prefetch(objset_t *os, uint64_t object, uint64_t offset,
-    uint64_t len);
+void dmu_prefetch(objset_t *os, uint64_t object, int64_t level, uint64_t offset,
+	uint64_t len, enum zio_priority pri);
 
 typedef struct dmu_object_info {
 	/* All sizes are in bytes unless otherwise indicated. */

include/sys/dsl_dataset.h

@@ -21,7 +21,7 @@
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2011, 2015 by Delphix. All rights reserved.
- * Copyright (c) 2013, Joyent, Inc. All rights reserved.
+ * Copyright (c) 2011, 2014 by Delphix. All rights reserved.
  * Copyright (c) 2013 Steven Hartland. All rights reserved.
  * Copyright (c) 2014 Spectra Logic Corporation, All rights reserved.
  */

include/sys/zfs_context.h

@@ -160,8 +160,18 @@ extern int aok;
 
 /*
  * DTrace SDT probes have different signatures in userland than they do in
- * kernel.  If they're being used in kernel code, re-define them out of
+ * the kernel.  If they're being used in kernel code, re-define them out of
  * existence for their counterparts in libzpool.
+ *
+ * Here's an example of how to use the set-error probes in userland:
+ * zfs$target:::set-error /arg0 == EBUSY/ {stack();}
+ *
+ * Here's an example of how to use DTRACE_PROBE probes in userland:
+ * If there is a probe declared as follows:
+ * DTRACE_PROBE2(zfs__probe_name, uint64_t, blkid, dnode_t *, dn);
+ * Then you can use it as follows:
+ * zfs$target:::probe2 /copyinstr(arg0) == "zfs__probe_name"/
+ *     {printf("%u %p\n", arg1, arg2);}
  */
 
 #ifdef DTRACE_PROBE