Illumos 4757, 4913

4757 ZFS embedded-data block pointers ("zero block compression")
4913 zfs release should not be subject to space checks

Reviewed by: Adam Leventhal <[email protected]>
Reviewed by: Max Grossman <[email protected]>
Reviewed by: George Wilson <[email protected]>
Reviewed by: Christopher Siden <[email protected]>
Reviewed by: Dan McDonald <[email protected]>
Approved by: Dan McDonald <[email protected]>

References:
  https://www.illumos.org/issues/4757
  https://www.illumos.org/issues/4913
  https://github.com/illumos/illumos-gate/commit/5d7b4d4

Porting notes:

For compatibility with the fastpath code the zio_done() function
needed to be updated.  Because embedded-data block pointers do
not require DVAs to be allocated the associated vdevs will not
be marked and therefore should not be unmarked.

Ported by: Tim Chase <[email protected]>
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #2544

3 files changed, 52 insertions, 7 deletions

diff --git a/lib/libzfs/libzfs_sendrecv.c b/lib/libzfs/libzfs_sendrecv.c
index a11d4bf36..ebd8a7be2 100644
--- a/lib/libzfs/libzfs_sendrecv.c
+++ b/lib/libzfs/libzfs_sendrecv.c
@@ -21,7 +21,7 @@
 
 /*
  * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
- * Copyright (c) 2013 by Delphix. All rights reserved.
+ * Copyright (c) 2011, 2014 by Delphix. All rights reserved.
  * Copyright (c) 2012, Joyent, Inc. All rights reserved.
  * Copyright (c) 2012 Pawel Jakub Dawidek <[email protected]>.
  * All rights reserved
@@ -49,6 +49,7 @@
 #include <time.h>
 
 #include <libzfs.h>
+#include <libzfs_core.h>
 
 #include "zfs_namecheck.h"
 #include "zfs_prop.h"
@@ -220,6 +221,7 @@ cksummer(void *arg)
 	struct drr_object *drro = &thedrr.drr_u.drr_object;
 	struct drr_write *drrw = &thedrr.drr_u.drr_write;
 	struct drr_spill *drrs = &thedrr.drr_u.drr_spill;
+	struct drr_write_embedded *drrwe = &thedrr.drr_u.drr_write_embedded;
 	FILE *ofp;
 	int outfd;
 	dmu_replay_record_t wbr_drr = {0};
@@ -415,6 +417,20 @@ cksummer(void *arg)
 			break;
 		}
 
+		case DRR_WRITE_EMBEDDED:
+		{
+			if (cksum_and_write(drr, sizeof (dmu_replay_record_t),
+			    &stream_cksum, outfd) == -1)
+				goto out;
+			(void) ssread(buf,
+			    P2ROUNDUP((uint64_t)drrwe->drr_psize, 8), ofp);
+			if (cksum_and_write(buf,
+			    P2ROUNDUP((uint64_t)drrwe->drr_psize, 8),
+			    &stream_cksum, outfd) == -1)
+				goto out;
+			break;
+		}
+
 		case DRR_FREE:
 		{
 			if (cksum_and_write(drr, sizeof (dmu_replay_record_t),
@@ -796,7 +812,7 @@ typedef struct send_dump_data {
 	char prevsnap[ZFS_MAXNAMELEN];
 	uint64_t prevsnap_obj;
 	boolean_t seenfrom, seento, replicate, doall, fromorigin;
-	boolean_t verbose, dryrun, parsable, progress;
+	boolean_t verbose, dryrun, parsable, progress, embed_data;
 	int outfd;
 	boolean_t err;
 	nvlist_t *fss;
@@ -876,7 +892,8 @@ estimate_ioctl(zfs_handle_t *zhp, uint64_t fromsnap_obj,
  */
 static int
 dump_ioctl(zfs_handle_t *zhp, const char *fromsnap, uint64_t fromsnap_obj,
-    boolean_t fromorigin, int outfd, nvlist_t *debugnv)
+    boolean_t fromorigin, int outfd, enum lzc_send_flags flags,
+    nvlist_t *debugnv)
 {
 	zfs_cmd_t zc = {"\0"};
 	libzfs_handle_t *hdl = zhp->zfs_hdl;
@@ -890,6 +907,7 @@ dump_ioctl(zfs_handle_t *zhp, const char *fromsnap, uint64_t fromsnap_obj,
 	zc.zc_obj = fromorigin;
 	zc.zc_sendobj = zfs_prop_get_int(zhp, ZFS_PROP_OBJSETID);
 	zc.zc_fromobj = fromsnap_obj;
+	zc.zc_flags = flags;
 
 	VERIFY(0 == nvlist_alloc(&thisdbg, NV_UNIQUE_NAME, 0));
 	if (fromsnap && fromsnap[0] != '\0') {
@@ -1140,8 +1158,12 @@ dump_snapshot(zfs_handle_t *zhp, void *arg)
 			}
 		}
 
+		enum lzc_send_flags flags = 0;
+		if (sdd->embed_data)
+			flags |= LZC_SEND_FLAG_EMBED_DATA;
+
 		err = dump_ioctl(zhp, sdd->prevsnap, sdd->prevsnap_obj,
-		    fromorigin, sdd->outfd, sdd->debugnv);
+		    fromorigin, sdd->outfd, flags, sdd->debugnv);
 
 		if (sdd->progress) {
 			(void) pthread_cancel(tid);
@@ -1485,6 +1507,7 @@ zfs_send(zfs_handle_t *zhp, const char *fromsnap, const char *tosnap,
 	sdd.parsable = flags->parsable;
 	sdd.progress = flags->progress;
 	sdd.dryrun = flags->dryrun;
+	sdd.embed_data = flags->embed_data;
 	sdd.filter_cb = filter_func;
 	sdd.filter_cb_arg = cb_arg;
 	if (debugnvp)
@@ -1616,7 +1639,8 @@ err_out:
 }
 
 int
-zfs_send_one(zfs_handle_t *zhp, const char *from, int fd)
+zfs_send_one(zfs_handle_t *zhp, const char *from, int fd,
+    enum lzc_send_flags flags)
 {
 	int err;
 	libzfs_handle_t *hdl = zhp->zfs_hdl;
@@ -1625,7 +1649,7 @@ zfs_send_one(zfs_handle_t *zhp, const char *from, int fd)
 	(void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN,
 	    "warning: cannot send '%s'"), zhp->zfs_name);
 
-	err = lzc_send(zhp->zfs_name, from, fd);
+	err = lzc_send(zhp->zfs_name, from, fd, flags);
 	if (err != 0) {
 		switch (errno) {
 		case EXDEV:
@@ -2543,6 +2567,16 @@ recv_skip(libzfs_handle_t *hdl, int fd, boolean_t byteswap)
 			(void) recv_read(hdl, fd, buf,
 			    drr->drr_u.drr_spill.drr_length, B_FALSE, NULL);
 			break;
+		case DRR_WRITE_EMBEDDED:
+			if (byteswap) {
+				drr->drr_u.drr_write_embedded.drr_psize =
+				    BSWAP_32(drr->drr_u.drr_write_embedded.
+				    drr_psize);
+			}
+			(void) recv_read(hdl, fd, buf,
+			    P2ROUNDUP(drr->drr_u.drr_write_embedded.drr_psize,
+			    8), B_FALSE, NULL);
+			break;
 		case DRR_WRITE_BYREF:
 		case DRR_FREEOBJECTS:
 		case DRR_FREE:
diff --git a/lib/libzfs_core/libzfs_core.c b/lib/libzfs_core/libzfs_core.c
index 0c05bbd27..2198ecd6f 100644
--- a/lib/libzfs_core/libzfs_core.c
+++ b/lib/libzfs_core/libzfs_core.c
@@ -439,6 +439,8 @@ lzc_get_holds(const char *snapname, nvlist_t **holdsp)
 }
 
 /*
+ * Generate a zfs send stream for the specified snapshot and write it to
+ * the specified file descriptor.
  *
  * "snapname" is the full name of the snapshot to send (e.g. "pool/fs@snap")
  *
@@ -452,9 +454,15 @@ lzc_get_holds(const char *snapname, nvlist_t **holdsp)
  * snapshot in the origin, etc.
  *
  * "fd" is the file descriptor to write the send stream to.
+ *
+ * If "flags" contains LZC_SEND_FLAG_EMBED_DATA, the stream is permitted
+ * to contain DRR_WRITE_EMBEDDED records with drr_etype==BP_EMBEDDED_TYPE_DATA,
+ * which the receiving system must support (as indicated by support
+ * for the "embedded_data" feature).
  */
 int
-lzc_send(const char *snapname, const char *from, int fd)
+lzc_send(const char *snapname, const char *from, int fd,
+    enum lzc_send_flags flags)
 {
 	nvlist_t *args;
 	int err;
@@ -463,6 +471,8 @@ lzc_send(const char *snapname, const char *from, int fd)
 	fnvlist_add_int32(args, "fd", fd);
 	if (from != NULL)
 		fnvlist_add_string(args, "fromsnap", from);
+	if (flags & LZC_SEND_FLAG_EMBED_DATA)
+		fnvlist_add_boolean(args, "embedok");
 	err = lzc_ioctl(ZFS_IOC_SEND_NEW, snapname, args, NULL);
 	nvlist_free(args);
 	return (err);
diff --git a/lib/libzpool/Makefile.am b/lib/libzpool/Makefile.am
index b960850fb..f4838da75 100644
--- a/lib/libzpool/Makefile.am
+++ b/lib/libzpool/Makefile.am
@@ -21,6 +21,7 @@ libzpool_la_SOURCES = \
 	$(top_srcdir)/module/zcommon/zpool_prop.c \
 	$(top_srcdir)/module/zcommon/zprop_common.c \
 	$(top_srcdir)/module/zfs/arc.c \
+	$(top_srcdir)/module/zfs/blkptr.c \
 	$(top_srcdir)/module/zfs/bplist.c \
 	$(top_srcdir)/module/zfs/bpobj.c \
 	$(top_srcdir)/module/zfs/bptree.c \