--- a/usr/src/cmd/Makefile	Mon Sep 21 21:03:13 2009 -0600
+++ b/usr/src/cmd/Makefile	Mon Sep 21 21:41:02 2009 -0600
@@ -461,6 +461,7 @@
 	zonename	\
 	zpool		\
 	zlook		\
+	zstreamdump	\
 	ztest
 
 $(CLOSED_BUILD)COMMON_SUBDIRS += \

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/usr/src/cmd/zstreamdump/Makefile	Mon Sep 21 21:41:02 2009 -0600
@@ -0,0 +1,67 @@
+#
+# CDDL HEADER START
+#
+# The contents of this file are subject to the terms of the
+# Common Development and Distribution License (the "License").
+# You may not use this file except in compliance with the License.
+#
+# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+# or http://www.opensolaris.org/os/licensing.
+# See the License for the specific language governing permissions
+# and limitations under the License.
+#
+# When distributing Covered Code, include this CDDL HEADER in each
+# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+# If applicable, add the following below this CDDL HEADER, with the
+# fields enclosed by brackets "[]" replaced with your own identifying
+# information: Portions Copyright [yyyy] [name of copyright owner]
+#
+# CDDL HEADER END
+#
+#
+# Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
+# Use is subject to license terms.
+#
+
+PROG=		zstreamdump
+OBJS=		zstreamdump.o
+SRCS=		$(OBJS:%.o=%.c)
+POFILE=		zstreamdump.po
+
+include ../Makefile.cmd
+
+INCS +=	-I../../uts/common/fs/zfs
+INCS +=	-I../../common/zfs
+
+LDLIBS += -lzfs -lnvpair
+
+C99MODE=	-xc99=%all
+C99LMODE=	-Xc99=%all
+
+CPPFLAGS += -D_LARGEFILE64_SOURCE=1 -D_REENTRANT $(INCS)
+$(NOT_RELEASE_BUILD)CPPFLAGS += -DDEBUG
+
+# lint complains about unused _umem_* functions
+LINTFLAGS += -xerroff=E_NAME_DEF_NOT_USED2 
+LINTFLAGS64 += -xerroff=E_NAME_DEF_NOT_USED2 
+
+.KEEP_STATE:
+
+.PARALLEL:
+
+all: $(PROG)
+
+$(PROG): $(OBJS)
+	$(LINK.c) -o $@ $(OBJS) $(LDLIBS)
+	$(POST_PROCESS)
+
+install: all $(ROOTUSRSBINPROG)
+
+clean:
+	$(RM) $(OBJS)
+
+lint:	lint_SRCS
+
+FRC:
+
+include ../Makefile.targ

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/usr/src/cmd/zstreamdump/zstreamdump.c	Mon Sep 21 21:41:02 2009 -0600
@@ -0,0 +1,363 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
+ *
+ * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+ * or http://www.opensolaris.org/os/licensing.
+ * See the License for the specific language governing permissions
+ * and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+
+/*
+ * Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
+ * Use is subject to license terms.
+ */
+
+#include <libnvpair.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <strings.h>
+#include <unistd.h>
+
+#include <sys/dmu.h>
+#include <sys/zfs_ioctl.h>
+#include <zfs_fletcher.h>
+
+uint64_t drr_record_count[DRR_NUMTYPES];
+uint64_t total_write_size = 0;
+uint64_t total_stream_len = 0;
+FILE *send_stream = 0;
+boolean_t do_byteswap = B_FALSE;
+boolean_t do_cksum = B_TRUE;
+#define	INITIAL_BUFLEN (1<<20)
+
+static void
+usage(void)
+{
+	(void) fprintf(stderr, "usage: zstreamdump [-v] [-C] < file\n");
+	(void) fprintf(stderr, "\t -v -- verbose\n");
+	(void) fprintf(stderr, "\t -C -- suppress checksum verification\n");
+	exit(1);
+}
+
+/*
+ * ssread - send stream read.
+ *
+ * Read while computing incremental checksum
+ */
+
+static size_t
+ssread(void *buf, size_t len, zio_cksum_t *cksum)
+{
+	size_t outlen;
+
+	if ((outlen = fread(buf, len, 1, send_stream)) == 0)
+		return (0);
+
+	if (do_cksum && cksum) {
+		if (do_byteswap)
+			fletcher_4_incremental_byteswap(buf, len, cksum);
+		else
+			fletcher_4_incremental_native(buf, len, cksum);
+	}
+	total_stream_len += len;
+	return (outlen);
+}
+
+int
+main(int argc, char *argv[])
+{
+	char *buf = malloc(INITIAL_BUFLEN);
+	dmu_replay_record_t thedrr;
+	dmu_replay_record_t *drr = &thedrr;
+	struct drr_begin *drrb = &thedrr.drr_u.drr_begin;
+	struct drr_end *drre = &thedrr.drr_u.drr_end;
+	struct drr_object *drro = &thedrr.drr_u.drr_object;
+	struct drr_freeobjects *drrfo = &thedrr.drr_u.drr_freeobjects;
+	struct drr_write *drrw = &thedrr.drr_u.drr_write;
+	struct drr_free *drrf = &thedrr.drr_u.drr_free;
+	char c;
+	boolean_t verbose = B_FALSE;
+	boolean_t first = B_TRUE;
+	int i, err;
+	zio_cksum_t zc = { 0 };
+	zio_cksum_t pcksum = { 0 };
+
+	while ((c = getopt(argc, argv, ":vC")) != -1) {
+		switch (c) {
+		case 'C':
+			do_cksum = B_FALSE;
+			break;
+		case 'v':
+			verbose = B_TRUE;
+			break;
+		case ':':
+			(void) fprintf(stderr,
+			    "missing argument for '%c' option\n", optopt);
+			usage();
+			break;
+		case '?':
+			(void) fprintf(stderr, "invalid option '%c'\n",
+			    optopt);
+			usage();
+		}
+	}
+
+	if (isatty(STDIN_FILENO)) {
+		(void) fprintf(stderr,
+		    "Error: Backup stream can not be read "
+		    "from a terminal.\n"
+		    "You must redirect standard input.\n");
+		exit(1);
+	}
+
+	send_stream = stdin;
+	pcksum = zc;
+	while (ssread(drr, sizeof (dmu_replay_record_t), &zc)) {
+
+		if (first) {
+			if (drrb->drr_magic == BSWAP_64(DMU_BACKUP_MAGIC)) {
+				do_byteswap = B_TRUE;
+				if (do_cksum) {
+					ZIO_SET_CHECKSUM(&zc, 0, 0, 0, 0);
+					/*
+					 * recalculate header checksum now
+					 * that we know it needs to be
+					 * byteswapped.
+					 */
+					fletcher_4_incremental_byteswap(drr,
+					    sizeof (dmu_replay_record_t), &zc);
+				}
+			} else if (drrb->drr_magic != DMU_BACKUP_MAGIC) {
+				(void) fprintf(stderr, "Invalid stream "
+				    "(bad magic number)\n");
+				exit(1);
+			}
+			first = B_FALSE;
+		}
+		if (do_byteswap) {
+			drr->drr_type = BSWAP_32(drr->drr_type);
+			drr->drr_payloadlen =
+			    BSWAP_32(drr->drr_payloadlen);
+		}
+
+		/*
+		 * At this point, the leading fields of the replay record
+		 * (drr_type and drr_payloadlen) have been byte-swapped if
+		 * necessary, but the rest of the data structure (the
+		 * union of type-specific structures) is still in its
+		 * original state.
+		 */
+		if (drr->drr_type >= DRR_NUMTYPES) {
+			(void) printf("INVALID record found: type 0x%x\n",
+			    drr->drr_type);
+			(void) printf("Aborting.\n");
+			exit(1);
+		}
+
+		drr_record_count[drr->drr_type]++;
+
+		switch (drr->drr_type) {
+		case DRR_BEGIN:
+			if (do_byteswap) {
+				drrb->drr_magic = BSWAP_64(drrb->drr_magic);
+				drrb->drr_version = BSWAP_64(drrb->drr_version);
+				drrb->drr_creation_time =
+				    BSWAP_64(drrb->drr_creation_time);
+				drrb->drr_type = BSWAP_32(drrb->drr_type);
+				drrb->drr_flags = BSWAP_32(drrb->drr_flags);
+				drrb->drr_toguid = BSWAP_64(drrb->drr_toguid);
+				drrb->drr_fromguid =
+				    BSWAP_64(drrb->drr_fromguid);
+			}
+
+			(void) printf("BEGIN record\n");
+			(void) printf("\tversion = %llx\n",
+			    (u_longlong_t)drrb->drr_version);
+			(void) printf("\tmagic = %llx\n",
+			    (u_longlong_t)drrb->drr_magic);
+			(void) printf("\tcreation_time = %llx\n",
+			    (u_longlong_t)drrb->drr_creation_time);
+			(void) printf("\ttype = %u\n", drrb->drr_type);
+			(void) printf("\tflags = 0x%x\n", drrb->drr_flags);
+			(void) printf("\ttoguid = %llx\n",
+			    (u_longlong_t)drrb->drr_toguid);
+			(void) printf("\tfromguid = %llx\n",
+			    (u_longlong_t)drrb->drr_fromguid);
+			(void) printf("\ttoname = %s\n", drrb->drr_toname);
+			if (verbose)
+				(void) printf("\n");
+
+			if (drrb->drr_version == 2 &&
+			    drr->drr_payloadlen != 0) {
+				nvlist_t *nv;
+				int sz = drr->drr_payloadlen;
+
+				if (sz > 1<<20) {
+					free(buf);
+					buf = malloc(sz);
+				}
+				(void) ssread(buf, sz, &zc);
+				if (ferror(send_stream))
+					perror("fread");
+				err = nvlist_unpack(buf, sz, &nv, 0);
+				if (err)
+					perror(strerror(err));
+				nvlist_print(stdout, nv);
+				nvlist_free(nv);
+			}
+			break;
+
+		case DRR_END:
+			if (do_byteswap) {
+				drre->drr_checksum.zc_word[0] =
+				    BSWAP_64(drre->drr_checksum.zc_word[0]);
+				drre->drr_checksum.zc_word[1] =
+				    BSWAP_64(drre->drr_checksum.zc_word[1]);
+				drre->drr_checksum.zc_word[2] =
+				    BSWAP_64(drre->drr_checksum.zc_word[2]);
+				drre->drr_checksum.zc_word[3] =
+				    BSWAP_64(drre->drr_checksum.zc_word[3]);
+			}
+			/*
+			 * We compare against the *previous* checksum
+			 * value, because the stored checksum is of
+			 * everything before the DRR_END record.
+			 */
+			if (do_cksum && !ZIO_CHECKSUM_EQUAL(drre->drr_checksum,
+			    pcksum)) {
+				(void) printf("Expected checksum differs from "
+				    "checksum in stream.\n");
+				(void) printf("Expected checksum = "
+				    "%llx/%llx/%llx/%llx\n",
+				    pcksum.zc_word[0],
+				    pcksum.zc_word[1],
+				    pcksum.zc_word[2],
+				    pcksum.zc_word[3]);
+			}
+			(void) printf("END checksum = %llx/%llx/%llx/%llx\n",
+			    drre->drr_checksum.zc_word[0],
+			    drre->drr_checksum.zc_word[1],
+			    drre->drr_checksum.zc_word[2],
+			    drre->drr_checksum.zc_word[3]);
+
+			ZIO_SET_CHECKSUM(&zc, 0, 0, 0, 0);
+			break;
+
+		case DRR_OBJECT:
+			if (do_byteswap) {
+				drro->drr_object = BSWAP_64(drro->drr_object);
+				drro->drr_type = BSWAP_32(drro->drr_type);
+				drro->drr_bonustype =
+				    BSWAP_32(drro->drr_bonustype);
+				drro->drr_blksz = BSWAP_32(drro->drr_blksz);
+				drro->drr_bonuslen =
+				    BSWAP_32(drro->drr_bonuslen);
+			}
+			if (verbose) {
+				(void) printf("OBJECT object = %llu type = %u "
+				    "bonustype = %u blksz = %u bonuslen = %u\n",
+				    (u_longlong_t)drro->drr_object,
+				    drro->drr_type,
+				    drro->drr_bonustype,
+				    drro->drr_blksz,
+				    drro->drr_bonuslen);
+			}
+			if (drro->drr_bonuslen > 0) {
+				(void) ssread(buf, P2ROUNDUP(drro->drr_bonuslen,
+				    8), &zc);
+			}
+			break;
+
+		case DRR_FREEOBJECTS:
+			if (do_byteswap) {
+				drrfo->drr_firstobj =
+				    BSWAP_64(drrfo->drr_firstobj);
+				drrfo->drr_numobjs =
+				    BSWAP_64(drrfo->drr_numobjs);
+			}
+			if (verbose) {
+				(void) printf("FREEOBJECTS firstobj = %llu "
+				    "numobjs = %llu\n",
+				    (u_longlong_t)drrfo->drr_firstobj,
+				    (u_longlong_t)drrfo->drr_numobjs);
+			}
+			break;
+
+		case DRR_WRITE:
+			if (do_byteswap) {
+				drrw->drr_object = BSWAP_64(drrw->drr_object);
+				drrw->drr_type = BSWAP_32(drrw->drr_type);
+				drrw->drr_offset = BSWAP_64(drrw->drr_offset);
+				drrw->drr_length = BSWAP_64(drrw->drr_length);
+			}
+			if (verbose) {
+				(void) printf("WRITE object = %llu type = %u "
+				    "offset = %llu length = %llu\n",
+				    (u_longlong_t)drrw->drr_object,
+				    drrw->drr_type,
+				    (u_longlong_t)drrw->drr_offset,
+				    (u_longlong_t)drrw->drr_length);
+			}
+			(void) ssread(buf, drrw->drr_length, &zc);
+			total_write_size += drrw->drr_length;
+			break;
+
+		case DRR_FREE:
+			if (do_byteswap) {
+				drrf->drr_object = BSWAP_64(drrf->drr_object);
+				drrf->drr_offset = BSWAP_64(drrf->drr_offset);
+				drrf->drr_length = BSWAP_64(drrf->drr_length);
+			}
+			if (verbose) {
+				(void) printf("FREE object = %llu "
+				    "offset = %llu length = %lld\n",
+				    (u_longlong_t)drrf->drr_object,
+				    (u_longlong_t)drrf->drr_offset,
+				    (longlong_t)drrf->drr_length);
+			}
+			break;
+		}
+		pcksum = zc;
+	}
+	free(buf);
+
+	/* Print final summary */
+
+	(void) printf("SUMMARY:\n");
+	(void) printf("\tTotal DRR_BEGIN records = %lld\n",
+	    (u_longlong_t)drr_record_count[DRR_BEGIN]);
+	(void) printf("\tTotal DRR_END records = %lld\n",
+	    (u_longlong_t)drr_record_count[DRR_END]);
+	(void) printf("\tTotal DRR_OBJECT records = %lld\n",
+	    (u_longlong_t)drr_record_count[DRR_OBJECT]);
+	(void) printf("\tTotal DRR_FREEOBJECTS records = %lld\n",
+	    (u_longlong_t)drr_record_count[DRR_FREEOBJECTS]);
+	(void) printf("\tTotal DRR_WRITE records = %lld\n",
+	    (u_longlong_t)drr_record_count[DRR_WRITE]);
+	(void) printf("\tTotal DRR_FREE records = %lld\n",
+	    (u_longlong_t)drr_record_count[DRR_FREE]);
+	(void) printf("\tTotal records = %lld\n",
+	    (u_longlong_t)(drr_record_count[DRR_BEGIN] +
+	    drr_record_count[DRR_OBJECT] +
+	    drr_record_count[DRR_FREEOBJECTS] +
+	    drr_record_count[DRR_WRITE] +
+	    drr_record_count[DRR_FREE] +
+	    drr_record_count[DRR_END]));
+	(void) printf("\tTotal write size = %lld (0x%llx)\n",
+	    (u_longlong_t)total_write_size, (u_longlong_t)total_write_size);
+	(void) printf("\tTotal stream length = %lld (0x%llx)\n",
+	    (u_longlong_t)total_stream_len, (u_longlong_t)total_stream_len);
+	return (0);
+}

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/usr/src/common/zfs/zfs_fletcher.c	Mon Sep 21 21:41:02 2009 -0600
@@ -0,0 +1,245 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
+ *
+ * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+ * or http://www.opensolaris.org/os/licensing.
+ * See the License for the specific language governing permissions
+ * and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+/*
+ * Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
+ * Use is subject to license terms.
+ */
+
+/*
+ * Fletcher Checksums
+ * ------------------
+ *
+ * ZFS's 2nd and 4th order Fletcher checksums are defined by the following
+ * recurrence relations:
+ *
+ *	a  = a    + f
+ *	 i    i-1    i-1
+ *
+ *	b  = b    + a
+ *	 i    i-1    i
+ *
+ *	c  = c    + b		(fletcher-4 only)
+ *	 i    i-1    i
+ *
+ *	d  = d    + c		(fletcher-4 only)
+ *	 i    i-1    i
+ *
+ * Where
+ *	a_0 = b_0 = c_0 = d_0 = 0
+ * and
+ *	f_0 .. f_(n-1) are the input data.
+ *
+ * Using standard techniques, these translate into the following series:
+ *
+ *	     __n_			     __n_
+ *	     \   |			     \   |
+ *	a  =  >     f			b  =  >     i * f
+ *	 n   /___|   n - i		 n   /___|	 n - i
+ *	     i = 1			     i = 1
+ *
+ *
+ *	     __n_			     __n_
+ *	     \   |  i*(i+1)		     \   |  i*(i+1)*(i+2)
+ *	c  =  >     ------- f		d  =  >     ------------- f
+ *	 n   /___|     2     n - i	 n   /___|	  6	   n - i
+ *	     i = 1			     i = 1
+ *
+ * For fletcher-2, the f_is are 64-bit, and [ab]_i are 64-bit accumulators.
+ * Since the additions are done mod (2^64), errors in the high bits may not
+ * be noticed.  For this reason, fletcher-2 is deprecated.
+ *
+ * For fletcher-4, the f_is are 32-bit, and [abcd]_i are 64-bit accumulators.
+ * A conservative estimate of how big the buffer can get before we overflow
+ * can be estimated using f_i = 0xffffffff for all i:
+ *
+ * % bc
+ *  f=2^32-1;d=0; for (i = 1; d<2^64; i++) { d += f*i*(i+1)*(i+2)/6 }; (i-1)*4
+ * 2264
+ *  quit
+ * %
+ *
+ * So blocks of up to 2k will not overflow.  Our largest block size is
+ * 128k, which has 32k 4-byte words, so we can compute the largest possible
+ * accumulators, then divide by 2^64 to figure the max amount of overflow:
+ *
+ * % bc
+ *  a=b=c=d=0; f=2^32-1; for (i=1; i<=32*1024; i++) { a+=f; b+=a; c+=b; d+=c }
+ *  a/2^64;b/2^64;c/2^64;d/2^64
+ * 0
+ * 0
+ * 1365
+ * 11186858
+ *  quit
+ * %
+ *
+ * So a and b cannot overflow.  To make sure each bit of input has some
+ * effect on the contents of c and d, we can look at what the factors of
+ * the coefficients in the equations for c_n and d_n are.  The number of 2s
+ * in the factors determines the lowest set bit in the multiplier.  Running
+ * through the cases for n*(n+1)/2 reveals that the highest power of 2 is
+ * 2^14, and for n*(n+1)*(n+2)/6 it is 2^15.  So while some data may overflow
+ * the 64-bit accumulators, every bit of every f_i effects every accumulator,
+ * even for 128k blocks.
+ *
+ * If we wanted to make a stronger version of fletcher4 (fletcher4c?),
+ * we could do our calculations mod (2^32 - 1) by adding in the carries
+ * periodically, and store the number of carries in the top 32-bits.
+ *
+ * --------------------
+ * Checksum Performance
+ * --------------------
+ *
+ * There are two interesting components to checksum performance: cached and
+ * uncached performance.  With cached data, fletcher-2 is about four times
+ * faster than fletcher-4.  With uncached data, the performance difference is
+ * negligible, since the cost of a cache fill dominates the processing time.
+ * Even though fletcher-4 is slower than fletcher-2, it is still a pretty
+ * efficient pass over the data.
+ *
+ * In normal operation, the data which is being checksummed is in a buffer
+ * which has been filled either by:
+ *
+ *	1. a compression step, which will be mostly cached, or
+ *	2. a bcopy() or copyin(), which will be uncached (because the
+ *	   copy is cache-bypassing).
+ *
+ * For both cached and uncached data, both fletcher checksums are much faster
+ * than sha-256, and slower than 'off', which doesn't touch the data at all.
+ */
+
+#include <sys/types.h>
+#include <sys/sysmacros.h>
+#include <sys/byteorder.h>
+#include <sys/spa.h>
+
+void
+fletcher_2_native(const void *buf, uint64_t size, zio_cksum_t *zcp)
+{
+	const uint64_t *ip = buf;
+	const uint64_t *ipend = ip + (size / sizeof (uint64_t));
+	uint64_t a0, b0, a1, b1;
+
+	for (a0 = b0 = a1 = b1 = 0; ip < ipend; ip += 2) {
+		a0 += ip[0];
+		a1 += ip[1];
+		b0 += a0;
+		b1 += a1;
+	}
+
+	ZIO_SET_CHECKSUM(zcp, a0, a1, b0, b1);
+}
+
+void
+fletcher_2_byteswap(const void *buf, uint64_t size, zio_cksum_t *zcp)
+{
+	const uint64_t *ip = buf;
+	const uint64_t *ipend = ip + (size / sizeof (uint64_t));
+	uint64_t a0, b0, a1, b1;
+
+	for (a0 = b0 = a1 = b1 = 0; ip < ipend; ip += 2) {
+		a0 += BSWAP_64(ip[0]);
+		a1 += BSWAP_64(ip[1]);
+		b0 += a0;
+		b1 += a1;
+	}
+
+	ZIO_SET_CHECKSUM(zcp, a0, a1, b0, b1);
+}
+
+void
+fletcher_4_native(const void *buf, uint64_t size, zio_cksum_t *zcp)
+{
+	const uint32_t *ip = buf;
+	const uint32_t *ipend = ip + (size / sizeof (uint32_t));
+	uint64_t a, b, c, d;
+
+	for (a = b = c = d = 0; ip < ipend; ip++) {
+		a += ip[0];
+		b += a;
+		c += b;
+		d += c;
+	}
+
+	ZIO_SET_CHECKSUM(zcp, a, b, c, d);
+}
+
+void
+fletcher_4_byteswap(const void *buf, uint64_t size, zio_cksum_t *zcp)
+{
+	const uint32_t *ip = buf;
+	const uint32_t *ipend = ip + (size / sizeof (uint32_t));
+	uint64_t a, b, c, d;
+
+	for (a = b = c = d = 0; ip < ipend; ip++) {
+		a += BSWAP_32(ip[0]);
+		b += a;
+		c += b;
+		d += c;
+	}
+
+	ZIO_SET_CHECKSUM(zcp, a, b, c, d);
+}
+
+void
+fletcher_4_incremental_native(const void *buf, uint64_t size,
+    zio_cksum_t *zcp)
+{
+	const uint32_t *ip = buf;
+	const uint32_t *ipend = ip + (size / sizeof (uint32_t));
+	uint64_t a, b, c, d;
+
+	a = zcp->zc_word[0];
+	b = zcp->zc_word[1];
+	c = zcp->zc_word[2];
+	d = zcp->zc_word[3];
+
+	for (; ip < ipend; ip++) {
+		a += ip[0];
+		b += a;
+		c += b;
+		d += c;
+	}
+
+	ZIO_SET_CHECKSUM(zcp, a, b, c, d);
+}
+
+void
+fletcher_4_incremental_byteswap(const void *buf, uint64_t size,
+    zio_cksum_t *zcp)
+{
+	const uint32_t *ip = buf;
+	const uint32_t *ipend = ip + (size / sizeof (uint32_t));
+	uint64_t a, b, c, d;
+
+	a = zcp->zc_word[0];
+	b = zcp->zc_word[1];
+	c = zcp->zc_word[2];
+	d = zcp->zc_word[3];
+
+	for (; ip < ipend; ip++) {
+		a += BSWAP_32(ip[0]);
+		b += a;
+		c += b;
+		d += c;
+	}
+
+	ZIO_SET_CHECKSUM(zcp, a, b, c, d);
+}

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/usr/src/common/zfs/zfs_fletcher.h	Mon Sep 21 21:41:02 2009 -0600
@@ -0,0 +1,53 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
+ *
+ * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+ * or http://www.opensolaris.org/os/licensing.
+ * See the License for the specific language governing permissions
+ * and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+/*
+ * Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
+ * Use is subject to license terms.
+ */
+
+#ifndef	_ZFS_FLETCHER_H
+#define	_ZFS_FLETCHER_H
+
+#include <sys/types.h>
+#include <sys/spa.h>
+
+#ifdef	__cplusplus
+extern "C" {
+#endif
+
+/*
+ * fletcher checksum functions
+ */
+
+void fletcher_2_native(const void *, uint64_t, zio_cksum_t *);
+void fletcher_2_byteswap(const void *, uint64_t, zio_cksum_t *);
+void fletcher_4_native(const void *, uint64_t, zio_cksum_t *);
+void fletcher_4_byteswap(const void *, uint64_t, zio_cksum_t *);
+void fletcher_4_incremental_native(const void *, uint64_t,
+    zio_cksum_t *);
+void fletcher_4_incremental_byteswap(const void *, uint64_t,
+    zio_cksum_t *);
+
+#ifdef	__cplusplus
+}
+#endif
+
+#endif	/* _ZFS_FLETCHER_H */

--- a/usr/src/lib/libzfs/Makefile.com	Mon Sep 21 21:03:13 2009 -0600
+++ b/usr/src/lib/libzfs/Makefile.com	Mon Sep 21 21:41:02 2009 -0600
@@ -27,7 +27,7 @@
 VERS= .1
 
 OBJS_SHARED= zfs_namecheck.o zprop_common.o zfs_prop.o zpool_prop.o \
-	zfs_deleg.o zfs_comutil.o
+	zfs_deleg.o zfs_comutil.o zfs_fletcher.o
 OBJS_COMMON= libzfs_dataset.o libzfs_util.o libzfs_graph.o libzfs_mount.o \
 	libzfs_pool.o libzfs_changelist.o libzfs_config.o libzfs_import.o \
 	libzfs_status.o libzfs_sendrecv.o

--- a/usr/src/lib/libzfs/common/libzfs_sendrecv.c	Mon Sep 21 21:03:13 2009 -0600
+++ b/usr/src/lib/libzfs/common/libzfs_sendrecv.c	Mon Sep 21 21:41:02 2009 -0600
@@ -40,10 +40,9 @@
 
 #include "zfs_namecheck.h"
 #include "zfs_prop.h"
+#include "zfs_fletcher.h"
 #include "libzfs_impl.h"
 
-#include <fletcher.c> /* XXX */
-
 static int zfs_receive_impl(libzfs_handle_t *, const char *, recvflags_t,
     int, avl_tree_t *, char **);
 

--- a/usr/src/lib/libzfs/common/llib-lzfs	Mon Sep 21 21:03:13 2009 -0600
+++ b/usr/src/lib/libzfs/common/llib-lzfs	Mon Sep 21 21:41:02 2009 -0600
@@ -19,14 +19,13 @@
  * CDDL HEADER END
  */
 /*
- * Copyright 2008 Sun Microsystems, Inc.  All rights reserved.
+ * Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
  * Use is subject to license terms.
  */
 
-#pragma ident	"%Z%%M%	%I%	%E% SMI"
-
 /*LINTLIBRARY*/
 /*PROTOLIB1*/
 
 #include <libzfs.h>
 #include "../../../common/zfs/zfs_comutil.h"
+#include "../../../common/zfs/zfs_fletcher.h"

--- a/usr/src/lib/libzfs/common/mapfile-vers	Mon Sep 21 21:03:13 2009 -0600
+++ b/usr/src/lib/libzfs/common/mapfile-vers	Mon Sep 21 21:41:02 2009 -0600
@@ -39,6 +39,12 @@
 
 SUNWprivate_1.1 {
     global:
+	fletcher_2_native;
+	fletcher_2_byteswap;
+	fletcher_4_native;
+	fletcher_4_byteswap;
+	fletcher_4_incremental_native;
+	fletcher_4_incremental_byteswap;
 	libzfs_errno;
 	libzfs_error_action;
 	libzfs_error_description;

--- a/usr/src/pkgdefs/SUNWzfsu/prototype_com	Mon Sep 21 21:03:13 2009 -0600
+++ b/usr/src/pkgdefs/SUNWzfsu/prototype_com	Mon Sep 21 21:41:02 2009 -0600
@@ -87,3 +87,4 @@
 l none usr/sbin/zdb=../../usr/lib/isaexec
 s none usr/sbin/zfs=../../sbin/zfs
 s none usr/sbin/zpool=../../sbin/zpool
+f none usr/sbin/zstreamdump 555 root bin

--- a/usr/src/uts/common/Makefile.files	Mon Sep 21 21:03:13 2009 -0600
+++ b/usr/src/uts/common/Makefile.files	Mon Sep 21 21:41:02 2009 -0600
@@ -1288,7 +1288,6 @@
 	dsl_deleg.o		\
 	dsl_prop.o		\
 	dsl_scrub.o		\
-	fletcher.o		\
 	gzip.o			\
 	lzjb.o			\
 	metaslab.o		\
@@ -1330,6 +1329,7 @@
 	zfs_deleg.o		\
 	zfs_prop.o		\
 	zfs_comutil.o		\
+	zfs_fletcher.o		\
 	zpool_prop.o		\
 	zprop_common.o
 

--- a/usr/src/uts/common/fs/zfs/fletcher.c	Mon Sep 21 21:03:13 2009 -0600
+++ /dev/null	Thu Jan 01 00:00:00 1970 +0000
@@ -1,245 +0,0 @@
-/*
- * CDDL HEADER START
- *
- * The contents of this file are subject to the terms of the
- * Common Development and Distribution License (the "License").
- * You may not use this file except in compliance with the License.
- *
- * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
- * or http://www.opensolaris.org/os/licensing.
- * See the License for the specific language governing permissions
- * and limitations under the License.
- *
- * When distributing Covered Code, include this CDDL HEADER in each
- * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
- * If applicable, add the following below this CDDL HEADER, with the
- * fields enclosed by brackets "[]" replaced with your own identifying
- * information: Portions Copyright [yyyy] [name of copyright owner]
- *
- * CDDL HEADER END
- */
-/*
- * Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
- * Use is subject to license terms.
- */
-
-/*
- * Fletcher Checksums
- * ------------------
- *
- * ZFS's 2nd and 4th order Fletcher checksums are defined by the following
- * recurrence relations:
- *
- *	a  = a    + f
- *	 i    i-1    i-1
- *
- *	b  = b    + a
- *	 i    i-1    i
- *
- *	c  = c    + b		(fletcher-4 only)
- *	 i    i-1    i
- *
- *	d  = d    + c		(fletcher-4 only)
- *	 i    i-1    i
- *
- * Where
- *	a_0 = b_0 = c_0 = d_0 = 0
- * and
- *	f_0 .. f_(n-1) are the input data.
- *
- * Using standard techniques, these translate into the following series:
- *
- *	     __n_			     __n_
- *	     \   |			     \   |
- *	a  =  >     f			b  =  >     i * f
- *	 n   /___|   n - i		 n   /___|	 n - i
- *	     i = 1			     i = 1
- *
- *
- *	     __n_			     __n_
- *	     \   |  i*(i+1)		     \   |  i*(i+1)*(i+2)
- *	c  =  >     ------- f		d  =  >     ------------- f
- *	 n   /___|     2     n - i	 n   /___|	  6	   n - i
- *	     i = 1			     i = 1
- *
- * For fletcher-2, the f_is are 64-bit, and [ab]_i are 64-bit accumulators.
- * Since the additions are done mod (2^64), errors in the high bits may not
- * be noticed.  For this reason, fletcher-2 is deprecated.
- *
- * For fletcher-4, the f_is are 32-bit, and [abcd]_i are 64-bit accumulators.
- * A conservative estimate of how big the buffer can get before we overflow
- * can be estimated using f_i = 0xffffffff for all i:
- *
- * % bc
- *  f=2^32-1;d=0; for (i = 1; d<2^64; i++) { d += f*i*(i+1)*(i+2)/6 }; (i-1)*4
- * 2264
- *  quit
- * %
- *
- * So blocks of up to 2k will not overflow.  Our largest block size is
- * 128k, which has 32k 4-byte words, so we can compute the largest possible
- * accumulators, then divide by 2^64 to figure the max amount of overflow:
- *
- * % bc
- *  a=b=c=d=0; f=2^32-1; for (i=1; i<=32*1024; i++) { a+=f; b+=a; c+=b; d+=c }
- *  a/2^64;b/2^64;c/2^64;d/2^64
- * 0
- * 0
- * 1365
- * 11186858
- *  quit
- * %
- *
- * So a and b cannot overflow.  To make sure each bit of input has some
- * effect on the contents of c and d, we can look at what the factors of
- * the coefficients in the equations for c_n and d_n are.  The number of 2s
- * in the factors determines the lowest set bit in the multiplier.  Running
- * through the cases for n*(n+1)/2 reveals that the highest power of 2 is
- * 2^14, and for n*(n+1)*(n+2)/6 it is 2^15.  So while some data may overflow
- * the 64-bit accumulators, every bit of every f_i effects every accumulator,
- * even for 128k blocks.
- *
- * If we wanted to make a stronger version of fletcher4 (fletcher4c?),
- * we could do our calculations mod (2^32 - 1) by adding in the carries
- * periodically, and store the number of carries in the top 32-bits.
- *
- * --------------------
- * Checksum Performance
- * --------------------
- *
- * There are two interesting components to checksum performance: cached and
- * uncached performance.  With cached data, fletcher-2 is about four times
- * faster than fletcher-4.  With uncached data, the performance difference is
- * negligible, since the cost of a cache fill dominates the processing time.
- * Even though fletcher-4 is slower than fletcher-2, it is still a pretty
- * efficient pass over the data.
- *
- * In normal operation, the data which is being checksummed is in a buffer
- * which has been filled either by:
- *
- *	1. a compression step, which will be mostly cached, or
- *	2. a bcopy() or copyin(), which will be uncached (because the
- *	   copy is cache-bypassing).
- *
- * For both cached and uncached data, both fletcher checksums are much faster
- * than sha-256, and slower than 'off', which doesn't touch the data at all.
- */
-
-#include <sys/types.h>
-#include <sys/sysmacros.h>
-#include <sys/byteorder.h>
-#include <sys/spa.h>
-
-void
-fletcher_2_native(const void *buf, uint64_t size, zio_cksum_t *zcp)
-{
-	const uint64_t *ip = buf;
-	const uint64_t *ipend = ip + (size / sizeof (uint64_t));
-	uint64_t a0, b0, a1, b1;
-
-	for (a0 = b0 = a1 = b1 = 0; ip < ipend; ip += 2) {
-		a0 += ip[0];
-		a1 += ip[1];
-		b0 += a0;
-		b1 += a1;
-	}
-
-	ZIO_SET_CHECKSUM(zcp, a0, a1, b0, b1);
-}
-
-void
-fletcher_2_byteswap(const void *buf, uint64_t size, zio_cksum_t *zcp)
-{
-	const uint64_t *ip = buf;
-	const uint64_t *ipend = ip + (size / sizeof (uint64_t));
-	uint64_t a0, b0, a1, b1;
-
-	for (a0 = b0 = a1 = b1 = 0; ip < ipend; ip += 2) {
-		a0 += BSWAP_64(ip[0]);
-		a1 += BSWAP_64(ip[1]);
-		b0 += a0;
-		b1 += a1;
-	}
-
-	ZIO_SET_CHECKSUM(zcp, a0, a1, b0, b1);
-}
-
-void
-fletcher_4_native(const void *buf, uint64_t size, zio_cksum_t *zcp)
-{
-	const uint32_t *ip = buf;
-	const uint32_t *ipend = ip + (size / sizeof (uint32_t));
-	uint64_t a, b, c, d;
-
-	for (a = b = c = d = 0; ip < ipend; ip++) {
-		a += ip[0];
-		b += a;
-		c += b;
-		d += c;
-	}
-
-	ZIO_SET_CHECKSUM(zcp, a, b, c, d);
-}
-
-void
-fletcher_4_byteswap(const void *buf, uint64_t size, zio_cksum_t *zcp)
-{
-	const uint32_t *ip = buf;
-	const uint32_t *ipend = ip + (size / sizeof (uint32_t));
-	uint64_t a, b, c, d;
-
-	for (a = b = c = d = 0; ip < ipend; ip++) {
-		a += BSWAP_32(ip[0]);
-		b += a;
-		c += b;
-		d += c;
-	}
-
-	ZIO_SET_CHECKSUM(zcp, a, b, c, d);
-}
-
-void
-fletcher_4_incremental_native(const void *buf, uint64_t size,
-    zio_cksum_t *zcp)
-{
-	const uint32_t *ip = buf;
-	const uint32_t *ipend = ip + (size / sizeof (uint32_t));
-	uint64_t a, b, c, d;
-
-	a = zcp->zc_word[0];
-	b = zcp->zc_word[1];
-	c = zcp->zc_word[2];
-	d = zcp->zc_word[3];
-
-	for (; ip < ipend; ip++) {
-		a += ip[0];
-		b += a;
-		c += b;
-		d += c;
-	}
-
-	ZIO_SET_CHECKSUM(zcp, a, b, c, d);
-}
-
-void
-fletcher_4_incremental_byteswap(const void *buf, uint64_t size,
-    zio_cksum_t *zcp)
-{
-	const uint32_t *ip = buf;
-	const uint32_t *ipend = ip + (size / sizeof (uint32_t));
-	uint64_t a, b, c, d;
-
-	a = zcp->zc_word[0];
-	b = zcp->zc_word[1];
-	c = zcp->zc_word[2];
-	d = zcp->zc_word[3];
-
-	for (; ip < ipend; ip++) {
-		a += BSWAP_32(ip[0]);
-		b += a;
-		c += b;
-		d += c;
-	}
-
-	ZIO_SET_CHECKSUM(zcp, a, b, c, d);
-}

--- a/usr/src/uts/common/fs/zfs/sys/zfs_ioctl.h	Mon Sep 21 21:03:13 2009 -0600
+++ b/usr/src/uts/common/fs/zfs/sys/zfs_ioctl.h	Mon Sep 21 21:41:02 2009 -0600
@@ -58,7 +58,7 @@
 typedef struct dmu_replay_record {
 	enum {
 		DRR_BEGIN, DRR_OBJECT, DRR_FREEOBJECTS,
-		DRR_WRITE, DRR_FREE, DRR_END,
+		DRR_WRITE, DRR_FREE, DRR_END, DRR_NUMTYPES
 	} drr_type;
 	uint32_t drr_payloadlen;
 	union {