diff options
42 files changed, 2763 insertions, 2719 deletions
@@ -20832,8 +20832,8 @@ f usr/xpg4/bin/basename 0555 root bin f usr/xpg4/bin/batch 0555 root bin s usr/xpg4/bin/bg=../../bin/alias s usr/xpg4/bin/cd=../../bin/alias -f usr/xpg4/bin/chgrp 0555 root bin -f usr/xpg4/bin/chown 0555 root bin +s usr/xpg4/bin/chgrp=../../bin/chgrp +s usr/xpg4/bin/chown=../../bin/chown s usr/xpg4/bin/command=../../bin/alias f usr/xpg4/bin/cp 0555 root bin f usr/xpg4/bin/crontab 4555 root bin diff --git a/usr/src/cmd/chgrp/Makefile b/usr/src/cmd/chgrp/Makefile index cac25266b7..7de84a25df 100644 --- a/usr/src/cmd/chgrp/Makefile +++ b/usr/src/cmd/chgrp/Makefile @@ -19,49 +19,37 @@ # CDDL HEADER END # # -#ident "%Z%%M% %I% %E% SMI" -# # Copyright 2008 Sun Microsystems, Inc. All rights reserved. # Use is subject to license terms. # PROG= chgrp -XPG4PROG= chgrp -XD= exobjs.xpg4 EXOBJS= chgrp.o -XPG4EXOBJS= exobjs.xpg4/chgrp.o include ../Makefile.cmd -$(XPG4) := CFLAGS += -DXPG4 +ROOTXPG4LINK=$(ROOTXPG4BIN)/$(PROG) + CPPFLAGS += -D_FILE_OFFSET_BITS=64 LDLIBS += -lcmdutils -lsec .KEEP_STATE: -all: $(PROG) $(XPG4) +all: $(PROG) $(PROG): $(EXOBJS) $(LINK.c) -o $@ $(EXOBJS) $(LDLIBS) $(POST_PROCESS) -$(XPG4): $(XD) $(XPG4EXOBJS) - $(LINK.c) -o $@ $(XPG4EXOBJS) $(LDLIBS) - $(POST_PROCESS) +$(ROOTXPG4LINK): $(ROOTPROG) + -$(RM) $@ + -$(SYMLINK) ../../bin/chgrp $@ -install: all $(ROOTPROG) $(ROOTXPG4PROG) +install: all $(ROOTPROG) $(ROOTXPG4LINK) clean: - -@rm -rf $(EXOBJS) $(XD) + -@rm -rf $(EXOBJS) lint: lint_PROG -$(XPG4EXOBJS): $(XD) - -$(XD)/%.o: %.c - $(COMPILE.c) -o $@ $< - -$(XD): - -@mkdir -p $@ - include ../Makefile.targ diff --git a/usr/src/cmd/chgrp/chgrp.c b/usr/src/cmd/chgrp/chgrp.c index a1fe7fefbf..ed27bbd47a 100644 --- a/usr/src/cmd/chgrp/chgrp.c +++ b/usr/src/cmd/chgrp/chgrp.c @@ -38,8 +38,6 @@ * contributors. */ -#pragma ident "%Z%%M% %I% %E% SMI" - /* * chgrp [-fhR] gid file ... * chgrp -R [-f] [-H|-L|-P] gid file ... @@ -82,41 +80,18 @@ static int isnumber(char *); static int Perror(char *); static void chgrpr(char *, gid_t); -#ifdef XPG4 -/* - * Check to see if we are to follow symlinks specified on the command line. - * This assumes we've already checked to make sure neither -h or -P was - * specified, so we are just looking to see if -R -L, or -R -H was specified, - * or, since -R has the same behavior as -R -L, if -R was specified by itself. - * Therefore, all we really need to check for is if -R was specified. - */ -#define FOLLOW_CL_LINKS (rflag) -#else /* * Check to see if we are to follow symlinks specified on the command line. * This assumes we've already checked to make sure neither -h or -P was * specified, so we are just looking to see if -R -L, or -R -H was specified. - * Note: -R by itself will change the group of a directory referenced by a - * symlink however it will not follow the symlink to any other part of the - * file hierarchy. */ #define FOLLOW_CL_LINKS (rflag && (Hflag || Lflag)) -#endif -#ifdef XPG4 -/* - * Follow symlinks when traversing directories. Since -R behaves the - * same as -R -L, we always want to follow symlinks to other parts - * of the file hierarchy unless -H was specified. - */ -#define FOLLOW_D_LINKS (!Hflag) -#else /* * Follow symlinks when traversing directories. Only follow symlinks * to other parts of the file hierarchy if -L was specified. */ #define FOLLOW_D_LINKS (Lflag) -#endif #define CHOWN(f, u, g) if (chown(f, u, g) < 0) { \ status += Perror(f); \ @@ -185,6 +160,13 @@ main(int argc, char *argv[]) default: usage(); } + /* + * Set Pflag by default for recursive operations + * if no other options were specified. + */ + if (rflag && !(Lflag || Hflag || Pflag || hflag)) { + Pflag = 1; + } /* * Check for sufficient arguments diff --git a/usr/src/cmd/chown/Makefile b/usr/src/cmd/chown/Makefile index 622030cf38..9ee61bfc5f 100644 --- a/usr/src/cmd/chown/Makefile +++ b/usr/src/cmd/chown/Makefile @@ -19,50 +19,37 @@ # CDDL HEADER END # # -#ident "%Z%%M% %I% %E% SMI" -# # Copyright 2008 Sun Microsystems, Inc. All rights reserved. # Use is subject to license terms. # PROG= chown -XPG4PROG= chown -XD= exobjs.xpg4 EXOBJS= chown.o -XPG4EXOBJS= exobjs.xpg4/chown.o include ../Makefile.cmd -$(XPG4) := CFLAGS += -DXPG4 +ROOTXPG4LINK=$(ROOTXPG4BIN)/$(PROG) + CPPFLAGS += -D_FILE_OFFSET_BITS=64 LDLIBS += -lcmdutils -lsec .KEEP_STATE: -all: $(PROG) $(XPG4) +all: $(PROG) $(PROG): $(EXOBJS) $(LINK.c) -o $@ $(EXOBJS) $(LDLIBS) $(POST_PROCESS) -$(XPG4): $(XD) $(XPG4EXOBJS) - $(LINK.c) -o $@ $(XPG4EXOBJS) $(LDLIBS) - $(POST_PROCESS) - +$(ROOTXPG4LINK): $(ROOTPROG) + -$(RM) $@ + -$(SYMLINK) ../../bin/chown $@ -install: all $(ROOTPROG) $(ROOTXPG4PROG) +install: all $(ROOTPROG) $(ROOTXPG4LINK) clean: - -@rm -rf $(EXOBJS) $(XD) + -@rm -rf $(EXOBJS) lint: lint_PROG -$(XPG4EXOBJS): $(XD) - -$(XD)/%.o: %.c - $(COMPILE.c) -o $@ $< - -$(XD): - -@mkdir -p $@ - include ../Makefile.targ diff --git a/usr/src/cmd/chown/chown.c b/usr/src/cmd/chown/chown.c index 319f571384..15db448a5b 100644 --- a/usr/src/cmd/chown/chown.c +++ b/usr/src/cmd/chown/chown.c @@ -24,15 +24,13 @@ */ /* Copyright (c) 1983, 1984, 1985, 1986, 1987, 1988, 1989 AT&T */ -/* All Rights Reserved */ +/* All Rights Reserved */ /* * Portions of this source code were derived from Berkeley 4.3 BSD * under license from the Regents of the University of California. */ -#pragma ident "%Z%%M% %I% %E% SMI" - /* * chown [-fhR] uid[:gid] file ... * chown -R [-f] [-H|-L|-P] uid[:gid] file ... @@ -76,41 +74,18 @@ static int isnumber(char *); static void chownr(char *, uid_t, gid_t); static void usage(); -#ifdef XPG4 -/* - * Check to see if we are to follow symlinks specified on the command line. - * This assumes we've already checked to make sure neither -h or -P was - * specified, so we are just looking to see if -R -H, or -R -L was specified, - * or, since -R has the same behavior as -R -L, if -R was specified by itself. - * Therefore, all we really need to check for is if -R was specified. - */ -#define FOLLOW_CL_LINKS (rflag) -#else /* * Check to see if we are to follow symlinks specified on the command line. * This assumes we've already checked to make sure neither -h or -P was * specified, so we are just looking to see if -R -H, or -R -L was specified. - * Note: -R by itself will change the ownership of a directory referenced by a - * symlink however it will now follow the symlink to any other part of the - * file hierarchy. */ #define FOLLOW_CL_LINKS (rflag && (Hflag || Lflag)) -#endif -#ifdef XPG4 -/* - * Follow symlinks when traversing directories. Since -R behaves the - * same as -R -L, we always want to follow symlinks to other parts - * of the file hierarchy unless -H was specified. - */ -#define FOLLOW_D_LINKS (!Hflag) -#else /* * Follow symlinks when traversing directories. Only follow symlinks * to other parts of the file hierarchy if -L was specified. */ #define FOLLOW_D_LINKS (Lflag) -#endif #define CHOWN(f, u, g) if (chown(f, u, g) < 0) { \ status += Perror(f); \ @@ -180,6 +155,13 @@ main(int argc, char *argv[]) } } /* + * Set Pflag by default for recursive operations + * if no other options were specified. + */ + if (rflag && !(Lflag || Hflag || Pflag || hflag)) { + Pflag = 1; + } + /* * Check for sufficient arguments * or a usage error. */ diff --git a/usr/src/cmd/cmd-inet/usr.bin/pppd/plugins/passprompt.c b/usr/src/cmd/cmd-inet/usr.bin/pppd/plugins/passprompt.c index ce361e9c37..a96fa32ba8 100644 --- a/usr/src/cmd/cmd-inet/usr.bin/pppd/plugins/passprompt.c +++ b/usr/src/cmd/cmd-inet/usr.bin/pppd/plugins/passprompt.c @@ -8,6 +8,11 @@ * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. */ + +/* + * Copyright (c) 2018, Joyent, Inc. + */ + #include <errno.h> #include <unistd.h> #include <fcntl.h> @@ -70,7 +75,7 @@ static int promptpass(char *user, char *passwd) argv[0] = promptprog; argv[1] = user == NULL ? "" : user; argv[2] = remote_name; - slprintf(fdstr, sizeof (fdstr), "%d", p[1]); + (void) slprintf(fdstr, sizeof (fdstr), "%d", p[1]); argv[3] = fdstr; argv[4] = NULL; (void) execv(*argv, argv); diff --git a/usr/src/cmd/cmd-inet/usr.sbin/ipadm/ipadm.c b/usr/src/cmd/cmd-inet/usr.sbin/ipadm/ipadm.c index 4dbbb83590..30e2f0f549 100644 --- a/usr/src/cmd/cmd-inet/usr.sbin/ipadm/ipadm.c +++ b/usr/src/cmd/cmd-inet/usr.sbin/ipadm/ipadm.c @@ -22,7 +22,7 @@ /* * Copyright (c) 2010, Oracle and/or its affiliates. All rights reserved. * Copyright 2017 Nexenta Systems, Inc. - * Copyright 2017 Joyent, Inc. + * Copyright (c) 2018, Joyent, Inc. * Copyright 2017 Gary Mills * Copyright (c) 2016, Chris Fraire <cfraire@me.com>. */ @@ -1636,8 +1636,7 @@ is_from_gz(const char *lifname) if (if_info->ifi_cflags & IFIF_L3PROTECT) ret = _B_TRUE; - if (if_info) - ipadm_free_if_info(if_info); + ipadm_free_if_info(if_info); return (ret); } diff --git a/usr/src/cmd/cmd-inet/usr.sbin/ping/ping.c b/usr/src/cmd/cmd-inet/usr.sbin/ping/ping.c index c6f8257ae9..9f959976a8 100644 --- a/usr/src/cmd/cmd-inet/usr.sbin/ping/ping.c +++ b/usr/src/cmd/cmd-inet/usr.sbin/ping/ping.c @@ -38,7 +38,7 @@ */ /* - * Copyright (c) 2017, Joyent, Inc. + * Copyright (c) 2018, Joyent, Inc. */ #include <assert.h> @@ -1989,7 +1989,7 @@ recv_icmp_packet(struct addrinfo *ai_dst, int recv_sock6, int recv_sock, progname, strerror(errno)); } continue; - } if (cc > 0) { + } else if (cc > 0) { check_reply(ai_dst, &in_msg, cc, udp_src_port); } diff --git a/usr/src/cmd/cmd-inet/usr.sbin/wificonfig/wificonfig.c b/usr/src/cmd/cmd-inet/usr.sbin/wificonfig/wificonfig.c index ad5100c1b0..5a9dc9889b 100644 --- a/usr/src/cmd/cmd-inet/usr.sbin/wificonfig/wificonfig.c +++ b/usr/src/cmd/cmd-inet/usr.sbin/wificonfig/wificonfig.c @@ -23,7 +23,9 @@ * Use is subject to license terms. */ -#pragma ident "%Z%%M% %I% %E% SMI" +/* + * Copyright (c) 2018, Joyent, Inc. + */ #include <stdio.h> #include <stdlib.h> @@ -2531,7 +2533,7 @@ do_deletepf(int fd, int argc, char **argv) B_FALSE); free(section_id); p_section = p_sectionbak; - continue; + continue; } p_section = p_section->section_next; } @@ -2751,8 +2753,9 @@ do_rmprefer(int fd, int argc, char **argv) return (B_FALSE); pae = plist->ael_head; while (pae != NULL) { + ae_t *next = pae->ae_next; free(pae); - pae = pae->ae_next; + pae = next; } plist->ael_head = plist->ael_tail = NULL; plist->ael_argc = 0; diff --git a/usr/src/cmd/ksh/builtins/alias.c b/usr/src/cmd/ksh/builtins/alias.c index a17e2dabd1..88edbc1e9b 100644 --- a/usr/src/cmd/ksh/builtins/alias.c +++ b/usr/src/cmd/ksh/builtins/alias.c @@ -68,9 +68,7 @@ bfastpathrec fastpath_builtins[] = /* This list must be alphabetically sorted for |strcmp()| usage */ { "basename", b_basename }, { "cat", b_cat }, - { "chgrp", b_chgrp }, { "chmod", b_chmod }, - { "chown", b_chown }, #ifdef ENABLE_PERFORMANCE_PARADOXON { "cksum", b_cksum }, #endif /* ENABLE_PERFORMANCE_PARADOXON */ @@ -98,7 +96,7 @@ bfastpathrec fastpath_builtins[] = { "mkfifo", b_mkfifo }, { "mktemp", b_mktemp }, { "mv", b_mv }, - { "paste", b_paste }, + { "paste", b_paste }, { "pathchk", b_pathchk }, { "pids", b_pids }, { "readlink", b_readlink }, @@ -119,7 +117,7 @@ bfastpathrec fastpath_builtins[] = { "uniq", b_uniq }, { "wc", b_wc }, { "xgrep", b_xgrep }, - { NULL, (int (*)(int, char **, void *))NULL } + { NULL, (int (*)(int, char **, void *))NULL } }; static inline diff --git a/usr/src/lib/libshell/common/data/solaris_cmdlist.h b/usr/src/lib/libshell/common/data/solaris_cmdlist.h index f394e79aa7..ba724a15b1 100644 --- a/usr/src/lib/libshell/common/data/solaris_cmdlist.h +++ b/usr/src/lib/libshell/common/data/solaris_cmdlist.h @@ -84,7 +84,7 @@ ASTCMDLIST(chgrp) ASTCMDLIST(chmod) ASTCMDLIST(chown) // XPG4CMDLIST(chown) -BINCMDLIST(chown) +// BINCMDLIST(chown) ASTCMDLIST(cksum) BINCMDLIST(cksum) GNUCMDLIST(cksum) diff --git a/usr/src/man/man1/chgrp.1 b/usr/src/man/man1/chgrp.1 index 795bc56f66..cab549e34b 100644 --- a/usr/src/man/man1/chgrp.1 +++ b/usr/src/man/man1/chgrp.1 @@ -49,7 +49,7 @@ .\" Portions Copyright (c) 1992, X/Open Company Limited All Rights Reserved .\" Copyright (c) 2003, Sun Microsystems, Inc. All Rights Reserved .\" -.TH CHGRP 1 "Jul 11, 2008" +.TH CHGRP 1 "Feb 21, 2019" .SH NAME chgrp \- change file group ownership .SH SYNOPSIS @@ -74,7 +74,6 @@ chgrp \- change file group ownership .fi .SH DESCRIPTION -.sp .LP The \fBchgrp\fR utility will set the group ID of the file named by each \fIfile\fR operand to the group ID specified by the \fIgroup\fR operand. @@ -136,10 +135,9 @@ set rstchown = 0 \fB_POSIX_CHOWN_RESTRICTED\fR is enabled by default. See \fBsystem\fR(4) and \fBfpathconf\fR(2). .SH OPTIONS -.sp .LP The following options are supported. -.SS "/usr/bin/chgrp and /usr/xpg4/bin/chgrp" + .sp .ne 2 .na @@ -202,49 +200,34 @@ of the file hierarchy. .sp .ne 2 .na -\fB\fB-s\fR\fR -.ad -.RS 6n -The specified group is Windows SID. This option requires a file system that -supports storing SIDs, such as ZFS. -.RE - -.sp -.LP -Specifying more than one of the mutually-exclusive options \fB-H\fR, \fB-L\fR, -or \fB-P\fR is not considered an error. The last option specified determines -the behavior of \fBchgrp\fR. -.SS "/usr/bin/chgrp" -.sp -.ne 2 -.na \fB\fB-R\fR\fR .ad .RS 6n Recursive. \fBchgrp\fR descends through the directory, and any subdirectories, setting the specified group \fBID\fR as it proceeds. When a symbolic link is -encountered, the group of the target file is changed, unless the \fB-h\fR or -\fB-P\fR option is specified. However, no recursion takes place, unless the -\fB-H\fR or \fB-L\fR option is specified. +encountered, the group of the of the symbolic link is changed, unless the +\fB-H\fR or \fB-L\fR option is specified. Unless the \fB-H\fR, \fB-L\fR, +or \fB-P\fR option is specified, the \fB-P\fR option is used as the default mode. .RE -.SS "/usr/xpg4/bin/chgrp" .sp .ne 2 .na -\fB\fB-R\fR\fR +\fB\fB-s\fR\fR .ad .RS 6n -Recursive. \fBchgrp\fR descends through the directory, and any subdirectories, -setting the specified group \fBID\fR as it proceeds. When a symbolic link is -encountered, the group of the target file is changed, unless the \fB-h\fR or -\fB-P\fR option is specified. Unless the \fB-H\fR, \fB-L\fR, or \fB-P\fR option -is specified, the \fB-L\fR option is used as the default mode. +The specified group is Windows SID. This option requires a file system that +supports storing SIDs, such as ZFS. .RE -.SH OPERANDS .sp .LP +Specifying more than one of the mutually-exclusive options \fB-H\fR, \fB-L\fR, +or \fB-P\fR is not considered an error. The last option specified determines +the behavior of \fBchgrp\fR. + +.SH OPERANDS +.LP The following operands are supported: .sp .ne 2 @@ -268,18 +251,15 @@ A path name of a file whose group ID is to be modified. .RE .SH USAGE -.sp .LP See \fBlargefile\fR(5) for the description of the behavior of \fBchgrp\fR when encountering files greater than or equal to 2 Gbyte (2^31 bytes). .SH ENVIRONMENT VARIABLES -.sp .LP See \fBenviron\fR(5) for descriptions of the following environment variables that affect the execution of \fBchgrp\fR: \fBLANG\fR, \fBLC_ALL\fR, \fBLC_CTYPE\fR, \fBLC_MESSAGES\fR, and \fBNLSPATH\fR. .SH EXIT STATUS -.sp .LP The following exit values are returned: .sp @@ -301,7 +281,6 @@ An error occurred. .RE .SH FILES -.sp .ne 2 .na \fB\fB/etc/group\fR\fR @@ -311,30 +290,9 @@ group file .RE .SH ATTRIBUTES -.sp .LP See \fBattributes\fR(5) for descriptions of the following attributes: -.SS "/usr/bin/chgrp" -.sp -.sp -.TS -box; -c | c -l | l . -ATTRIBUTE TYPE ATTRIBUTE VALUE -_ -CSI Enabled. See NOTES. -_ -Interface Stability Committed -_ -Standard See \fBstandards\fR(5). -.TE - -.SS "/usr/xpg4/bin/chgrp" -.sp - -.sp .TS box; c | c @@ -349,12 +307,14 @@ Standard See \fBstandards\fR(5). .TE .SH SEE ALSO -.sp .LP \fBchmod\fR(1), \fBchown\fR(1), \fBid\fR(1M), \fBchown\fR(2), \fBfpathconf\fR(2), \fBgroup\fR(4), \fBpasswd\fR(4), \fBsystem\fR(4), \fBattributes\fR(5), \fBenviron\fR(5), \fBlargefile\fR(5), \fBstandards\fR(5) .SH NOTES -.sp .LP \fBchgrp\fR is CSI-enabled except for the \fIgroup\fR name. +.sp +.LP +In the past the behavior of \fB/usr/xpg4/bin/chgrp\fR and +\fB/usr/bin/chgrp\fR utilities was different. Now they behave the same way. diff --git a/usr/src/man/man1/chown.1 b/usr/src/man/man1/chown.1 index f309fa3e37..6ec6a8247a 100644 --- a/usr/src/man/man1/chown.1 +++ b/usr/src/man/man1/chown.1 @@ -45,61 +45,31 @@ .\" Portions Copyright (c) 1982-2007 AT&T Knowledge Ventures .\" Copyright (c) 2008, Sun Microsystems, Inc. All Rights Reserved .\" -.TH CHOWN 1 "Jul 11, 2008" +.TH CHOWN 1 "Feb 21, 2019" .SH NAME chown \- change file ownership .SH SYNOPSIS -.SS "/usr/bin/chown" .LP .nf -\fB/usr/bin/chown\fR [\fB-fhR\fR] \fIowner\fR[:\fIgroup\fR] \fIfile\fR... +\fBchown\fR [\fB-fhR\fR] \fIowner\fR[:\fIgroup\fR] \fIfile\fR... .fi .LP .nf -\fB/usr/bin/chown\fR \fB-s\fR [\fB-fhR\fR] \fIownersid\fR[:\fIgroupsid\fR] \fIfile\fR... +\fBchown\fR \fB-s\fR [\fB-fhR\fR] \fIownersid\fR[:\fIgroupsid\fR] \fIfile\fR... .fi .LP .nf -\fB/usr/bin/chown\fR \fB-R\fR [\fB-f\fR] [\fB-H\fR | \fB-L\fR | \fB-P\fR] \fIowner\fR[:\fIgroup\fR] \fIfile\fR... +\fBchown\fR \fB-R\fR [\fB-f\fR] [\fB-H\fR | \fB-L\fR | \fB-P\fR] \fIowner\fR[:\fIgroup\fR] \fIfile\fR... .fi .LP .nf -\fB/usr/bin/chown\fR \fB-s\fR \fB-R\fR [\fB-f\fR] [\fB-H\fR | \fB-L\fR | \fB-P\fR] \fIownersid\fR[:\fIgroupsid\fR] \fIfile\fR... -.fi - -.SS "/usr/xpg4/bin/chown" -.LP -.nf -\fB/usr/xpg4/bin/chown\fR [\fB-fhR\fR] \fIowner\fR[:\fIgroup\fR] \fIfile\fR... -.fi - -.LP -.nf -\fB/usr/xpg4/bin/chown\fR \fB-s\fR [\fB-fhR\fR] \fIownersid\fR[:\fIgroupsid\fR] \fIfile\fR... -.fi - -.LP -.nf -\fB/usr/xpg4/bin/chown\fR \fB-R\fR [\fB-f\fR] [\fB-H\fR | \fB-L\fR | \fB-P\fR] \fIowner\fR[:\fIgroup\fR] \fIfile\fR... -.fi - -.LP -.nf -\fB/usr/xpg4/bin/chown\fR \fB-s\fR \fB-R\fR [\fB-f\fR] [\fB-H\fR | \fB-L\fR | \fB-P\fR] \fIownersid\fR[:\fIgroupsid\fR] \fIfile\fR... -.fi - -.SS "ksh93" -.LP -.nf -\fBchown\fR [\fB-cflhmnvHLPRX\fR] [\fB-r\fR \fIfile\fR] \fIowner\fR[:\fIgroup\fR] \fIfile\fR... +\fBchown\fR \fB-s\fR \fB-R\fR [\fB-f\fR] [\fB-H\fR | \fB-L\fR | \fB-P\fR] \fIownersid\fR[:\fIgroupsid\fR] \fIfile\fR... .fi .SH DESCRIPTION -.SS "/usr/bin/chown and /usr/xpg4/bin/chown" -.sp .LP The \fBchown\fR utility sets the user \fBID\fR of the file named by each \fBfile\fR to the user \fBID\fR specified by \fIowner\fR, and, optionally, sets @@ -142,22 +112,7 @@ To disable this option, include the following line in \fB/etc/system\fR: .LP \fB{_POSIX_CHOWN_RESTRICTED}\fR is enabled by default. See \fBsystem\fR(4) and \fBfpathconf\fR(2). -.SS "ksh93" -.sp -.LP -The \fBchown\fR built-in in \fBksh93\fR is associated with the \fB/bin\fR and -\fB/usr/bin\fR paths. It is invoked when \fBchown\fR is executed without a -pathname prefix and the pathname search finds a \fB/bin/chown\fR or -/usr/bin/chown executable. -.sp -.LP -\fBchown\fR changes the ownership of each file to \fIowner\fR. \fIowner\fR can -be specified as either a user name or a numeric user id. The group ownership of -each file can also be changed to \fIgroup\fR by appending \fI:group\fR to the -user name. .SH OPTIONS -.SS "/usr/bin/chown and /usr/xpg4/bin/chown" -.sp .LP The following options are supported: .sp @@ -222,239 +177,33 @@ of the file hierarchy. .sp .ne 2 .na -\fB\fB-s\fR\fR -.ad -.RS 6n -The owner and/or group arguments are Windows SID strings. This option requires -a file system that supports storing SIDs, such as ZFS. -.RE - -.sp -.LP -Specifying more than one of the mutually-exclusive options \fB-H\fR, \fB-L\fR, -or \fB-P\fR is not considered an error. The last option specified determines -the behavior of \fBchown\fR. -.SS "/usr/bin/chown" -.sp -.LP -The following options are supported: -.sp -.ne 2 -.na \fB\fB-R\fR\fR .ad .RS 6n Recursive. \fBchown\fR descends through the directory, and any subdirectories, setting the specified ownership \fBID\fR as it proceeds. When a symbolic link -is encountered, the owner of the target file is changed, unless the \fB-h\fR or -\fB-P\fR option is specified. However, no recursion takes place, unless the -\fB-H\fR or \fB-L\fR option is specified. +is encountered, the owner of the symbolic link is changed, unless the +\fB-H\fR or \fB-L\fR option is specified. Unless the \fB-H\fR, \fB-L\fR, or \fB-P\fR +option is specified, the \fB-P\fR option is used as the default mode. .RE -.SS "/usr/xpg4/bin/chown" -.sp -.LP -The following options are supported: .sp .ne 2 .na -\fB\fB-R\fR\fR +\fB\fB-s\fR\fR .ad .RS 6n -Recursive. \fBchown\fR descends through the directory, and any subdirectories, -setting the specified ownership \fBID\fR as it proceeds. When a symbolic link -is encountered, the owner of the target file is changed, unless the \fB-h\fR or -\fB-P\fR option is specified. Unless the \fB-H\fR, \fB-L\fR, or \fB-P\fR option -is specified, the \fB-L\fR option is used as the default mode. +The owner and/or group arguments are Windows SID strings. This option requires +a file system that supports storing SIDs, such as ZFS. .RE -.SS "ksh93" .sp .LP -The following options are supported by the \fBksh93\fR built-in \fBchown\fR -command: -.sp -.ne 2 -.na -\fB\fB-c\fR\fR -.ad -.br -.na -\fB\fB--changes\fR\fR -.ad -.sp .6 -.RS 4n -Describe only files whose ownership actually changes. -.RE - -.sp -.ne 2 -.na -\fB\fB-f\fR\fR -.ad -.br -.na -\fB\fB--quiet | silent\fR\fR -.ad -.sp .6 -.RS 4n -Do not report files whose ownership fails to change. -.RE - -.sp -.ne 2 -.na -\fB\fB-l | h\fR\fR -.ad -.br -.na -\fB\fB--symlink\fR\fR -.ad -.sp .6 -.RS 4n -Change the ownership of the symbolic links on systems that support this option. -.RE - -.sp -.ne 2 -.na -\fB\fB-m\fR\fR -.ad -.br -.na -\fB\fB--map\fR\fR -.ad -.sp .6 -.RS 4n -Interpret the first operand as a file that contains a map of: -.sp -.in +2 -.nf -\fIfrom_uid\fR:\fIfrom_gid to_uid:to_gid\fR -.fi -.in -2 -.sp - -pairs. Ownership of files matching the \fIfrom\fR part of any pair is changed -to the corresponding \fIto\fR part of the pair. The process stops at the first -match for each file. Unmatched files are silently ignored. -.RE - -.sp -.ne 2 -.na -\fB\fB-n\fR\fR -.ad -.br -.na -\fB\fB--show\fR\fR -.ad -.sp .6 -.RS 4n -Show actions but do not execute. -.RE - -.sp -.ne 2 -.na -\fB\fB-r\fR\fR -.ad -.br -.na -\fB\fB--reference=file\fR\fR -.ad -.sp .6 -.RS 4n -Omit the explicit ownership operand and use the ownership of the file instead. -.RE - -.sp -.ne 2 -.na -\fB\fB-v\fR\fR -.ad -.br -.na -\fB\fB--verbose\fR\fR -.ad -.sp .6 -.RS 4n -Describe the changed permissions of all files. -.RE - -.sp -.ne 2 -.na -\fB\fB-H\fR\fR -.ad -.br -.na -\fB\fB--metaphysical\fR\fR -.ad -.sp .6 -.RS 4n -Follow symbolic links for command arguments. Otherwise do not follow symbolic -links when traversing directories. -.RE - -.sp -.ne 2 -.na -\fB\fB-L\fR\fR -.ad -.br -.na -\fB\fB--logical | follow\fR\fR -.ad -.sp .6 -.RS 4n -Follow symbolic links when traversing directories. -.RE - -.sp -.ne 2 -.na -\fB\fB-P\fR\fR -.ad -.br -.na -\fB\fB--physical | nofollow\fR\fR -.ad -.sp .6 -.RS 4n -Do not follow symbolic links when traversing directories. -.RE - -.sp -.ne 2 -.na -\fB\fB-R\fR\fR -.ad -.br -.na -\fB\fB--recursive\fR\fR -.ad -.sp .6 -.RS 4n -Recursively change ownership of directories and their contents. -.RE - -.sp -.ne 2 -.na -\fB\fB-X\fR\fR -.ad -.br -.na -\fB\fB--test\fR\fR -.ad -.sp .6 -.RS 4n -Canonicalize output for testing. -.RE +Specifying more than one of the mutually-exclusive options \fB-H\fR, \fB-L\fR, +or \fB-P\fR is not considered an error. The last option specified determines +the behavior of \fBchown\fR. .SH OPERANDS -.sp .LP The following operands are supported: .sp @@ -486,7 +235,6 @@ A path name of a file whose user \fBID\fR is to be modified. .RE .SH USAGE -.sp .LP See \fBlargefile\fR(5) for the description of the behavior of \fBchown\fR when encountering files greater than or equal to 2 Gbyte ( 2^31 bytes). @@ -507,13 +255,11 @@ example% \fBchown \(miR \(mih \fIowner\fR[:group] \fIfile\fR...\fR .sp .SH ENVIRONMENT VARIABLES -.sp .LP See \fBenviron\fR(5) for descriptions of the following environment variables that affect the execution of \fBchown\fR: \fBLANG\fR, \fBLC_ALL\fR, \fBLC_CTYPE\fR, \fBLC_MESSAGES\fR, and \fBNLSPATH\fR. .SH EXIT STATUS -.sp .LP The following exit values are returned: .sp @@ -535,7 +281,6 @@ An error occurred. .RE .SH FILES -.sp .ne 2 .na \fB\fB/etc/passwd\fR\fR @@ -545,28 +290,8 @@ System password file .RE .SH ATTRIBUTES -.sp .LP See \fBattributes\fR(5) for descriptions of the following attributes: -.SS "/usr/bin/chown" -.sp - -.sp -.TS -box; -c | c -l | l . -ATTRIBUTE TYPE ATTRIBUTE VALUE -_ -CSI Enabled. See NOTES. -_ -Interface Stability Committed -_ -Standard See \fBstandards\fR(5). -.TE - -.SS "/usr/xpg4/bin/chown" -.sp .sp .TS @@ -582,31 +307,16 @@ _ Standard See \fBstandards\fR(5). .TE -.SS "ksh93" -.sp - -.sp -.TS -box; -c | c -l | l . -ATTRIBUTE TYPE ATTRIBUTE VALUE -_ -Interface Stability See below. -.TE - -.sp -.LP -The \fBksh93\fR built-in binding to \fB/bin\fR and \fB/usr/bin\fR is Volatile. -The built-in interfaces are Uncommitted. .SH SEE ALSO -.sp .LP \fBchgrp\fR(1), \fBchmod\fR(1), \fBksh93\fR(1), \fBchown\fR(2), \fBfpathconf\fR(2), \fBpasswd\fR(4), \fBsystem\fR(4), \fBattributes\fR(5), \fBenviron\fR(5), \fBlargefile\fR(5), \fBstandards\fR(5) .SH NOTES -.sp .LP \fBchown\fR is \fBCSI\fR-enabled except for the \fIowner\fR and \fIgroup\fR names. +.sp +.LP +In the past the behavior of \fB/usr/xpg4/bin/chown\fR and +\fB/usr/bin/chown\fR utilities was different. Now they behave the same way. diff --git a/usr/src/pkg/manifests/system-test-utiltest.mf b/usr/src/pkg/manifests/system-test-utiltest.mf index 5fdd891b11..3ed9baeba5 100644 --- a/usr/src/pkg/manifests/system-test-utiltest.mf +++ b/usr/src/pkg/manifests/system-test-utiltest.mf @@ -1118,6 +1118,7 @@ file path=opt/util-tests/tests/awk/tests/T.redir mode=0555 file path=opt/util-tests/tests/awk/tests/T.split mode=0555 file path=opt/util-tests/tests/awk/tests/T.sub mode=0555 file path=opt/util-tests/tests/awk/tests/T.system mode=0555 +file path=opt/util-tests/tests/chown_test mode=0555 file path=opt/util-tests/tests/date_test mode=0555 file path=opt/util-tests/tests/demangle/afl-fast mode=0555 file path=opt/util-tests/tests/demangle/gcc-libstdc++ mode=0555 @@ -1291,6 +1292,18 @@ file path=opt/util-tests/tests/dis/risc-v/tst.supervisor.out mode=0444 file path=opt/util-tests/tests/dis/risc-v/tst.supervisor.s mode=0444 file path=opt/util-tests/tests/dis/sparc/tst.regs.out mode=0444 file path=opt/util-tests/tests/dis/sparc/tst.regs.s mode=0444 +file path=opt/util-tests/tests/files/cout0 mode=0444 +file path=opt/util-tests/tests/files/cout1 mode=0444 +file path=opt/util-tests/tests/files/cout10 mode=0444 +file path=opt/util-tests/tests/files/cout11 mode=0444 +file path=opt/util-tests/tests/files/cout2 mode=0444 +file path=opt/util-tests/tests/files/cout3 mode=0444 +file path=opt/util-tests/tests/files/cout4 mode=0444 +file path=opt/util-tests/tests/files/cout5 mode=0444 +file path=opt/util-tests/tests/files/cout6 mode=0444 +file path=opt/util-tests/tests/files/cout7 mode=0444 +file path=opt/util-tests/tests/files/cout8 mode=0444 +file path=opt/util-tests/tests/files/cout9 mode=0444 file path=opt/util-tests/tests/files/gout0 mode=0444 file path=opt/util-tests/tests/files/gout1 mode=0444 file path=opt/util-tests/tests/files/gout10 mode=0444 diff --git a/usr/src/pkg/manifests/system-xopen-xcu4.mf b/usr/src/pkg/manifests/system-xopen-xcu4.mf index c9de2bb7fa..b7f5733eb4 100644 --- a/usr/src/pkg/manifests/system-xopen-xcu4.mf +++ b/usr/src/pkg/manifests/system-xopen-xcu4.mf @@ -38,8 +38,6 @@ file path=usr/xpg4/bin/at group=sys mode=4755 file path=usr/xpg4/bin/awk mode=0555 file path=usr/xpg4/bin/basename mode=0555 file path=usr/xpg4/bin/batch mode=0555 -file path=usr/xpg4/bin/chgrp mode=0555 -file path=usr/xpg4/bin/chown mode=0555 file path=usr/xpg4/bin/cp mode=0555 file path=usr/xpg4/bin/crontab mode=4555 file path=usr/xpg4/bin/ctags mode=0555 @@ -83,6 +81,8 @@ license lic_OSBL_preamble license=lic_OSBL_preamble link path=usr/xpg4/bin/alias target=../../bin/alias link path=usr/xpg4/bin/bg target=../../bin/alias link path=usr/xpg4/bin/cd target=../../bin/alias +link path=usr/xpg4/bin/chgrp target=../../bin/chgrp +link path=usr/xpg4/bin/chown target=../../bin/chown link path=usr/xpg4/bin/command target=../../bin/alias link path=usr/xpg4/bin/df target=../../sbin/df link path=usr/xpg4/bin/egrep target=../../bin/grep diff --git a/usr/src/test/util-tests/runfiles/default.run b/usr/src/test/util-tests/runfiles/default.run index b328670f8e..3cccf62ac9 100644 --- a/usr/src/test/util-tests/runfiles/default.run +++ b/usr/src/test/util-tests/runfiles/default.run @@ -52,6 +52,7 @@ tests = [ 'runtests.sh' ] [/opt/util-tests/tests/grep_test] [/opt/util-tests/tests/date_test] +[/opt/util-tests/tests/chown_test] [/opt/util-tests/tests/demangle] tests = ['afl-fast', 'gcc-libstdc++', 'llvm-stdcxxabi'] diff --git a/usr/src/test/util-tests/tests/Makefile b/usr/src/test/util-tests/tests/Makefile index f947f4dc71..c84899363e 100644 --- a/usr/src/test/util-tests/tests/Makefile +++ b/usr/src/test/util-tests/tests/Makefile @@ -19,6 +19,7 @@ SUBDIRS = date dis dladm iconv libnvpair_json libsff printf xargs grep_xpg4 SUBDIRS += demangle mergeq workq +SUBDIRS += demangle mergeq workq chown SUBDIRS += bunyan awk smbios libjedec include $(SRC)/test/Makefile.com diff --git a/usr/src/test/util-tests/tests/chown/Makefile b/usr/src/test/util-tests/tests/chown/Makefile new file mode 100644 index 0000000000..52c70f8b3c --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/Makefile @@ -0,0 +1,45 @@ +# +# 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. +# + +# +# Copyright (c) 2013 by Delphix. All rights reserved. +# Copyright 2014 Nexenta Systems, Inc. All rights reserved. +# Copyright 2019 Alexander Pyhalov +# + +.PARALLEL: $(SUBDIRS) + +include $(SRC)/cmd/Makefile.cmd +include $(SRC)/test/Makefile.com + +ROOTOPTPKG = $(ROOT)/opt/util-tests +TESTDIR = $(ROOTOPTPKG)/tests + +PROGS = chown_test + +CMDS = $(PROGS:%=$(TESTDIR)/%) +$(CMDS) := FILEMODE = 0555 + +all lint clean clobber: + +install: $(CMDS) + +$(CMDS): $(TESTDIR) + +$(TESTDIR): + $(INS.dir) + +$(TESTDIR)/%: %.ksh + $(INS.rename) + +SUBDIRS = files + +include $(SRC)/test/Makefile.com diff --git a/usr/src/test/util-tests/tests/chown/chown_test.ksh b/usr/src/test/util-tests/tests/chown/chown_test.ksh new file mode 100644 index 0000000000..d3f3478a9d --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/chown_test.ksh @@ -0,0 +1,103 @@ +#! /usr/bin/ksh +# +# +# This file and its contents are supplied under the terms of the +# Common Development and Distribution License ("CDDL"), version 1.0. +# You may only use this file in accordance with the terms of version +# 1.0 of the CDDL. +# +# A full copy of the text of the CDDL should have accompanied this +# source. A copy of the CDDL is also available via the Internet at +# http://www.illumos.org/license/CDDL. +# + +# +# Copyright 2017 Nexenta Systems, Inc. All rights reserved. +# Copyright 2019 Alexander Pyhalov +# + +CHOWN=${CHOWN:=/usr/bin/chown} +FILEDIR=/opt/util-tests/tests/files + +fail() { + echo $1 + exit -1 +} + +create_test_hier() { + mkdir -p $1/src $1/dst + touch $1/target + touch $1/file + touch $1/dst/file1 + touch $1/src/file2 + ln -s ../target $1/src/tlink + ln -s ../dst $1/src/dstlink + ln -s target $1/tlink +} + +SUCCESSFLAGS=" +-f +-h +-fhR +-R +-RH +-RP +-RL +-Rh +-RHL +-RPH +-RLP" + +FAILFLAGS="-RPh +-RLh +-RHh +-P +-H +-L" + +NEWOWNER=daemon + +# We set PATH to /bin to try get ksh chown builtin +# and to ensure that /usr/bin/chown is used instead. +export PATH=/bin + +# We want unified output from tools +export LC_ALL=en_US.UTF-8 + +i=0 +echo "$SUCCESSFLAGS" | while read flags; do + print -n "test $i: chown $flags: " + TD=$(mktemp -d -t) + if [ -d "$TD" ]; then + create_test_hier $TD + chown $flags $NEWOWNER $TD/src || fail "chown $flags $NEWOWNER $TD/src failed on exit" + chown $flags $NEWOWNER $TD/tlink || fail "chown $flags $NEWOWNER $TD/tlink failed on exit" + chown $flags $NEWOWNER $TD/file || fail "chown $flags $NEWOWNER $TD/file failed on exit" + (cd $TD ; find . -ls |\ + awk ' { print $3 " " $5 " " $11 }'|\ + sort -k 3 > /tmp/out.$$) + if [ -n "$(diff /tmp/out.$$ $FILEDIR/cout$i)" ]; then + print "$(diff -u /tmp/out.$$ $FILEDIR/cout$i)" + fail "result is different" + fi + echo "passed" + rm -fr $TD /tmp/out.$$ + else + fail "couldn't create $TD" + fi + ((i++)) +done + +echo "$FAILFLAGS" | while read flags; do + print -n "test $i: chown $flags: " + TD=$(mktemp -d -t) + if [ -d "$TD" ]; then + create_test_hier $TD + chown $flags $NEWOWNER $TD/file && fail "chown $flags $NEWOWNER $TD/file should have failed" + echo "passed" + rm -fr $TD + else + fail "couldn't create $TD" + fi + ((i++)) +done diff --git a/usr/src/test/util-tests/tests/chown/files/Makefile b/usr/src/test/util-tests/tests/chown/files/Makefile new file mode 100644 index 0000000000..4c46abaa63 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/Makefile @@ -0,0 +1,50 @@ +# +# 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. +# + +# +# Copyright (c) 2013 by Delphix. All rights reserved. +# Copyright 2014 Nexenta Systems, Inc. All rights reserved. +# Copyright 2019 Alexander Pyhalov +# + +include $(SRC)/cmd/Makefile.cmd +include $(SRC)/test/Makefile.com + +ROOTOPTPKG = $(ROOT)/opt/util-tests +TESTDIR = $(ROOTOPTPKG)/tests/files + +PROGS = cout0 \ + cout1 \ + cout2 \ + cout3 \ + cout4 \ + cout5 \ + cout6 \ + cout7 \ + cout8 \ + cout9 \ + cout10 \ + cout11 + +CMDS = $(PROGS:%=$(TESTDIR)/%) +$(CMDS) := FILEMODE = 0444 + +all lint clean clobber: + +install: $(CMDS) + +$(CMDS): $(TESTDIR) + +$(TESTDIR): + $(INS.dir) + +$(TESTDIR)/%: % + $(INS.file) diff --git a/usr/src/test/util-tests/tests/chown/files/cout0 b/usr/src/test/util-tests/tests/chown/files/cout0 new file mode 100644 index 0000000000..d12a3e0a90 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout0 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x root ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx root ./src/dstlink +-rw-r--r-- root ./src/file2 +lrwxrwxrwx root ./src/tlink +-rw-r--r-- daemon ./target +lrwxrwxrwx root ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout1 b/usr/src/test/util-tests/tests/chown/files/cout1 new file mode 100644 index 0000000000..d12a3e0a90 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout1 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x root ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx root ./src/dstlink +-rw-r--r-- root ./src/file2 +lrwxrwxrwx root ./src/tlink +-rw-r--r-- daemon ./target +lrwxrwxrwx root ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout10 b/usr/src/test/util-tests/tests/chown/files/cout10 new file mode 100644 index 0000000000..f128e53d0a --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout10 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x daemon ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx root ./src/dstlink +-rw-r--r-- daemon ./src/file2 +lrwxrwxrwx root ./src/tlink +-rw-r--r-- daemon ./target +lrwxrwxrwx root ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout11 b/usr/src/test/util-tests/tests/chown/files/cout11 new file mode 100644 index 0000000000..e990d2c3d7 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout11 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x root ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx daemon ./src/dstlink +-rw-r--r-- daemon ./src/file2 +lrwxrwxrwx daemon ./src/tlink +-rw-r--r-- root ./target +lrwxrwxrwx daemon ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout12 b/usr/src/test/util-tests/tests/chown/files/cout12 new file mode 100644 index 0000000000..e69de29bb2 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout12 diff --git a/usr/src/test/util-tests/tests/chown/files/cout2 b/usr/src/test/util-tests/tests/chown/files/cout2 new file mode 100644 index 0000000000..68b3cfaf8b --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout2 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x root ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx root ./src/dstlink +-rw-r--r-- root ./src/file2 +lrwxrwxrwx root ./src/tlink +-rw-r--r-- root ./target +lrwxrwxrwx daemon ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout3 b/usr/src/test/util-tests/tests/chown/files/cout3 new file mode 100644 index 0000000000..e990d2c3d7 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout3 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x root ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx daemon ./src/dstlink +-rw-r--r-- daemon ./src/file2 +lrwxrwxrwx daemon ./src/tlink +-rw-r--r-- root ./target +lrwxrwxrwx daemon ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout4 b/usr/src/test/util-tests/tests/chown/files/cout4 new file mode 100644 index 0000000000..e990d2c3d7 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout4 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x root ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx daemon ./src/dstlink +-rw-r--r-- daemon ./src/file2 +lrwxrwxrwx daemon ./src/tlink +-rw-r--r-- root ./target +lrwxrwxrwx daemon ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout5 b/usr/src/test/util-tests/tests/chown/files/cout5 new file mode 100644 index 0000000000..f128e53d0a --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout5 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x daemon ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx root ./src/dstlink +-rw-r--r-- daemon ./src/file2 +lrwxrwxrwx root ./src/tlink +-rw-r--r-- daemon ./target +lrwxrwxrwx root ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout6 b/usr/src/test/util-tests/tests/chown/files/cout6 new file mode 100644 index 0000000000..e990d2c3d7 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout6 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x root ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx daemon ./src/dstlink +-rw-r--r-- daemon ./src/file2 +lrwxrwxrwx daemon ./src/tlink +-rw-r--r-- root ./target +lrwxrwxrwx daemon ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout7 b/usr/src/test/util-tests/tests/chown/files/cout7 new file mode 100644 index 0000000000..dcf411a1e5 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout7 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x daemon ./dst +-rw-r--r-- daemon ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx root ./src/dstlink +-rw-r--r-- daemon ./src/file2 +lrwxrwxrwx root ./src/tlink +-rw-r--r-- daemon ./target +lrwxrwxrwx root ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout8 b/usr/src/test/util-tests/tests/chown/files/cout8 new file mode 100644 index 0000000000..e990d2c3d7 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout8 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x root ./dst +-rw-r--r-- root ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx daemon ./src/dstlink +-rw-r--r-- daemon ./src/file2 +lrwxrwxrwx daemon ./src/tlink +-rw-r--r-- root ./target +lrwxrwxrwx daemon ./tlink diff --git a/usr/src/test/util-tests/tests/chown/files/cout9 b/usr/src/test/util-tests/tests/chown/files/cout9 new file mode 100644 index 0000000000..dcf411a1e5 --- /dev/null +++ b/usr/src/test/util-tests/tests/chown/files/cout9 @@ -0,0 +1,10 @@ +drwx------ root . +drwxr-xr-x daemon ./dst +-rw-r--r-- daemon ./dst/file1 +-rw-r--r-- daemon ./file +drwxr-xr-x daemon ./src +lrwxrwxrwx root ./src/dstlink +-rw-r--r-- daemon ./src/file2 +lrwxrwxrwx root ./src/tlink +-rw-r--r-- daemon ./target +lrwxrwxrwx root ./tlink diff --git a/usr/src/test/zfs-tests/tests/functional/snapshot/deadlist_lock.ksh b/usr/src/test/zfs-tests/tests/functional/snapshot/deadlist_lock.ksh index 47f912dc9c..a592916a61 100644 --- a/usr/src/test/zfs-tests/tests/functional/snapshot/deadlist_lock.ksh +++ b/usr/src/test/zfs-tests/tests/functional/snapshot/deadlist_lock.ksh @@ -13,6 +13,7 @@ # # Copyright (c) 2016 by Delphix. All rights reserved. +# Copyright 2019 Joyent, Inc. # . $STF_SUITE/include/libtest.shlib @@ -43,15 +44,15 @@ # snapshot right after you create a new test file. # 2. Start DTrace in the background to put a delay in the # sync thread after it closes the empty bpobj and before -# it reopens it. +# it reopens it. The dtrace process is set to exit when this +# script exits. # 3. Start a process in the backgroud that runs zfs-destroy # dry-runs in an infinite loop. The idea is to keep calling # dsl_deadlist_space_range(). # 4. Go ahead and start removing the test files. This should # start populating the deadlist of each snapshot with # entries and go through the dle_enqueue() target code. -# 5. If the test passes, kill the process running on a loop -# and dtrace, and cleanup the dataset. +# 5. Kill the 'zfs destroy' loop and clean up the dataset. # verify_runnable "both" @@ -61,7 +62,7 @@ DLDS="dl_race" function cleanup { - log_must kill -9 $DLOOP_PID $DTRACE_PID + log_must kill -9 $DLOOP_PID log_must zfs destroy -fR $TESTPOOL/$TESTFS/$DLDS } @@ -86,8 +87,7 @@ log_onexit cleanup setup log_must sync -log_must dtrace -qwn "fbt::bpobj_decr_empty:entry { chill(500000000); }" & -DTRACE_PID="$!" +log_must dtrace -p "$PPID" -qwn "fbt::bpobj_decr_empty:entry { chill(500000000); }" & sleep 1 destroy_nv_loop & @@ -100,5 +100,5 @@ done log_must sync log_pass "There should be no race condition when an administrative command" \ - " attempts to read a deadlist's entries at the same time a that a sync" \ + " attempts to read a deadlist's entries while a sync" \ " thread is manipulating it." diff --git a/usr/src/uts/common/Makefile.files b/usr/src/uts/common/Makefile.files index 0c734ff062..8b236155fb 100644 --- a/usr/src/uts/common/Makefile.files +++ b/usr/src/uts/common/Makefile.files @@ -25,7 +25,7 @@ # Copyright (c) 2013 by Saso Kiselkov. All rights reserved. # Copyright 2018 Nexenta Systems, Inc. # Copyright 2016 Garrett D'Amore <garrett@damore.org> -# Copyright 2017 Joyent, Inc. +# Copyright 2019 Joyent, Inc. # Copyright 2016 OmniTI Computer Consulting, Inc. All rights reserved. # Copyright 2016 Hans Rosenfeld <rosenfeld@grumpf.hope-2000.org> # @@ -1356,6 +1356,7 @@ ZFS_COMMON_OBJS += \ ddt_zap.o \ dmu.o \ dmu_diff.o \ + dmu_recv.o \ dmu_send.o \ dmu_object.o \ dmu_objset.o \ diff --git a/usr/src/uts/common/fs/zfs/dmu_recv.c b/usr/src/uts/common/fs/zfs/dmu_recv.c new file mode 100644 index 0000000000..bee41bd95e --- /dev/null +++ b/usr/src/uts/common/fs/zfs/dmu_recv.c @@ -0,0 +1,2264 @@ +/* + * 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 (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. + * Copyright 2011 Nexenta Systems, Inc. All rights reserved. + * Copyright (c) 2011, 2015 by Delphix. All rights reserved. + * Copyright (c) 2014, Joyent, Inc. All rights reserved. + * Copyright 2014 HybridCluster. All rights reserved. + * Copyright 2016 RackTop Systems. + * Copyright (c) 2014 Integros [integros.com] + */ + +#include <sys/dmu.h> +#include <sys/dmu_impl.h> +#include <sys/dmu_tx.h> +#include <sys/dbuf.h> +#include <sys/dnode.h> +#include <sys/zfs_context.h> +#include <sys/dmu_objset.h> +#include <sys/dmu_traverse.h> +#include <sys/dsl_dataset.h> +#include <sys/dsl_dir.h> +#include <sys/dsl_prop.h> +#include <sys/dsl_pool.h> +#include <sys/dsl_synctask.h> +#include <sys/zfs_ioctl.h> +#include <sys/zap.h> +#include <sys/zio_checksum.h> +#include <sys/zfs_znode.h> +#include <zfs_fletcher.h> +#include <sys/avl.h> +#include <sys/ddt.h> +#include <sys/zfs_onexit.h> +#include <sys/dmu_recv.h> +#include <sys/dsl_destroy.h> +#include <sys/blkptr.h> +#include <sys/dsl_bookmark.h> +#include <sys/zfeature.h> +#include <sys/bqueue.h> + +int zfs_recv_queue_length = SPA_MAXBLOCKSIZE; + +static char *dmu_recv_tag = "dmu_recv_tag"; +const char *recv_clone_name = "%recv"; + +static void byteswap_record(dmu_replay_record_t *drr); + +typedef struct dmu_recv_begin_arg { + const char *drba_origin; + dmu_recv_cookie_t *drba_cookie; + cred_t *drba_cred; + uint64_t drba_snapobj; +} dmu_recv_begin_arg_t; + +static int +recv_begin_check_existing_impl(dmu_recv_begin_arg_t *drba, dsl_dataset_t *ds, + uint64_t fromguid) +{ + uint64_t val; + int error; + dsl_pool_t *dp = ds->ds_dir->dd_pool; + + /* temporary clone name must not exist */ + error = zap_lookup(dp->dp_meta_objset, + dsl_dir_phys(ds->ds_dir)->dd_child_dir_zapobj, recv_clone_name, + 8, 1, &val); + if (error != ENOENT) + return (error == 0 ? EBUSY : error); + + /* new snapshot name must not exist */ + error = zap_lookup(dp->dp_meta_objset, + dsl_dataset_phys(ds)->ds_snapnames_zapobj, + drba->drba_cookie->drc_tosnap, 8, 1, &val); + if (error != ENOENT) + return (error == 0 ? EEXIST : error); + + /* + * Check snapshot limit before receiving. We'll recheck again at the + * end, but might as well abort before receiving if we're already over + * the limit. + * + * Note that we do not check the file system limit with + * dsl_dir_fscount_check because the temporary %clones don't count + * against that limit. + */ + error = dsl_fs_ss_limit_check(ds->ds_dir, 1, ZFS_PROP_SNAPSHOT_LIMIT, + NULL, drba->drba_cred); + if (error != 0) + return (error); + + if (fromguid != 0) { + dsl_dataset_t *snap; + uint64_t obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; + + /* Find snapshot in this dir that matches fromguid. */ + while (obj != 0) { + error = dsl_dataset_hold_obj(dp, obj, FTAG, + &snap); + if (error != 0) + return (SET_ERROR(ENODEV)); + if (snap->ds_dir != ds->ds_dir) { + dsl_dataset_rele(snap, FTAG); + return (SET_ERROR(ENODEV)); + } + if (dsl_dataset_phys(snap)->ds_guid == fromguid) + break; + obj = dsl_dataset_phys(snap)->ds_prev_snap_obj; + dsl_dataset_rele(snap, FTAG); + } + if (obj == 0) + return (SET_ERROR(ENODEV)); + + if (drba->drba_cookie->drc_force) { + drba->drba_snapobj = obj; + } else { + /* + * If we are not forcing, there must be no + * changes since fromsnap. + */ + if (dsl_dataset_modified_since_snap(ds, snap)) { + dsl_dataset_rele(snap, FTAG); + return (SET_ERROR(ETXTBSY)); + } + drba->drba_snapobj = ds->ds_prev->ds_object; + } + + dsl_dataset_rele(snap, FTAG); + } else { + /* if full, then must be forced */ + if (!drba->drba_cookie->drc_force) + return (SET_ERROR(EEXIST)); + /* start from $ORIGIN@$ORIGIN, if supported */ + drba->drba_snapobj = dp->dp_origin_snap != NULL ? + dp->dp_origin_snap->ds_object : 0; + } + + return (0); + +} + +static int +dmu_recv_begin_check(void *arg, dmu_tx_t *tx) +{ + dmu_recv_begin_arg_t *drba = arg; + dsl_pool_t *dp = dmu_tx_pool(tx); + struct drr_begin *drrb = drba->drba_cookie->drc_drrb; + uint64_t fromguid = drrb->drr_fromguid; + int flags = drrb->drr_flags; + int error; + uint64_t featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo); + dsl_dataset_t *ds; + const char *tofs = drba->drba_cookie->drc_tofs; + + /* already checked */ + ASSERT3U(drrb->drr_magic, ==, DMU_BACKUP_MAGIC); + ASSERT(!(featureflags & DMU_BACKUP_FEATURE_RESUMING)); + + if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == + DMU_COMPOUNDSTREAM || + drrb->drr_type >= DMU_OST_NUMTYPES || + ((flags & DRR_FLAG_CLONE) && drba->drba_origin == NULL)) + return (SET_ERROR(EINVAL)); + + /* Verify pool version supports SA if SA_SPILL feature set */ + if ((featureflags & DMU_BACKUP_FEATURE_SA_SPILL) && + spa_version(dp->dp_spa) < SPA_VERSION_SA) + return (SET_ERROR(ENOTSUP)); + + if (drba->drba_cookie->drc_resumable && + !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EXTENSIBLE_DATASET)) + return (SET_ERROR(ENOTSUP)); + + /* + * The receiving code doesn't know how to translate a WRITE_EMBEDDED + * record to a plain WRITE record, so the pool must have the + * EMBEDDED_DATA feature enabled if the stream has WRITE_EMBEDDED + * records. Same with WRITE_EMBEDDED records that use LZ4 compression. + */ + if ((featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) && + !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EMBEDDED_DATA)) + return (SET_ERROR(ENOTSUP)); + if ((featureflags & DMU_BACKUP_FEATURE_LZ4) && + !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LZ4_COMPRESS)) + return (SET_ERROR(ENOTSUP)); + + /* + * The receiving code doesn't know how to translate large blocks + * to smaller ones, so the pool must have the LARGE_BLOCKS + * feature enabled if the stream has LARGE_BLOCKS. Same with + * large dnodes. + */ + if ((featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) && + !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_BLOCKS)) + return (SET_ERROR(ENOTSUP)); + if ((featureflags & DMU_BACKUP_FEATURE_LARGE_DNODE) && + !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_DNODE)) + return (SET_ERROR(ENOTSUP)); + + error = dsl_dataset_hold(dp, tofs, FTAG, &ds); + if (error == 0) { + /* target fs already exists; recv into temp clone */ + + /* Can't recv a clone into an existing fs */ + if (flags & DRR_FLAG_CLONE || drba->drba_origin) { + dsl_dataset_rele(ds, FTAG); + return (SET_ERROR(EINVAL)); + } + + error = recv_begin_check_existing_impl(drba, ds, fromguid); + dsl_dataset_rele(ds, FTAG); + } else if (error == ENOENT) { + /* target fs does not exist; must be a full backup or clone */ + char buf[ZFS_MAX_DATASET_NAME_LEN]; + + /* + * If it's a non-clone incremental, we are missing the + * target fs, so fail the recv. + */ + if (fromguid != 0 && !(flags & DRR_FLAG_CLONE || + drba->drba_origin)) + return (SET_ERROR(ENOENT)); + + /* + * If we're receiving a full send as a clone, and it doesn't + * contain all the necessary free records and freeobject + * records, reject it. + */ + if (fromguid == 0 && drba->drba_origin && + !(flags & DRR_FLAG_FREERECORDS)) + return (SET_ERROR(EINVAL)); + + /* Open the parent of tofs */ + ASSERT3U(strlen(tofs), <, sizeof (buf)); + (void) strlcpy(buf, tofs, strrchr(tofs, '/') - tofs + 1); + error = dsl_dataset_hold(dp, buf, FTAG, &ds); + if (error != 0) + return (error); + + /* + * Check filesystem and snapshot limits before receiving. We'll + * recheck snapshot limits again at the end (we create the + * filesystems and increment those counts during begin_sync). + */ + error = dsl_fs_ss_limit_check(ds->ds_dir, 1, + ZFS_PROP_FILESYSTEM_LIMIT, NULL, drba->drba_cred); + if (error != 0) { + dsl_dataset_rele(ds, FTAG); + return (error); + } + + error = dsl_fs_ss_limit_check(ds->ds_dir, 1, + ZFS_PROP_SNAPSHOT_LIMIT, NULL, drba->drba_cred); + if (error != 0) { + dsl_dataset_rele(ds, FTAG); + return (error); + } + + if (drba->drba_origin != NULL) { + dsl_dataset_t *origin; + error = dsl_dataset_hold(dp, drba->drba_origin, + FTAG, &origin); + if (error != 0) { + dsl_dataset_rele(ds, FTAG); + return (error); + } + if (!origin->ds_is_snapshot) { + dsl_dataset_rele(origin, FTAG); + dsl_dataset_rele(ds, FTAG); + return (SET_ERROR(EINVAL)); + } + if (dsl_dataset_phys(origin)->ds_guid != fromguid && + fromguid != 0) { + dsl_dataset_rele(origin, FTAG); + dsl_dataset_rele(ds, FTAG); + return (SET_ERROR(ENODEV)); + } + dsl_dataset_rele(origin, FTAG); + } + dsl_dataset_rele(ds, FTAG); + error = 0; + } + return (error); +} + +static void +dmu_recv_begin_sync(void *arg, dmu_tx_t *tx) +{ + dmu_recv_begin_arg_t *drba = arg; + dsl_pool_t *dp = dmu_tx_pool(tx); + objset_t *mos = dp->dp_meta_objset; + struct drr_begin *drrb = drba->drba_cookie->drc_drrb; + const char *tofs = drba->drba_cookie->drc_tofs; + dsl_dataset_t *ds, *newds; + uint64_t dsobj; + int error; + uint64_t crflags = 0; + + if (drrb->drr_flags & DRR_FLAG_CI_DATA) + crflags |= DS_FLAG_CI_DATASET; + + error = dsl_dataset_hold(dp, tofs, FTAG, &ds); + if (error == 0) { + /* create temporary clone */ + dsl_dataset_t *snap = NULL; + if (drba->drba_snapobj != 0) { + VERIFY0(dsl_dataset_hold_obj(dp, + drba->drba_snapobj, FTAG, &snap)); + } + dsobj = dsl_dataset_create_sync(ds->ds_dir, recv_clone_name, + snap, crflags, drba->drba_cred, tx); + if (drba->drba_snapobj != 0) + dsl_dataset_rele(snap, FTAG); + dsl_dataset_rele(ds, FTAG); + } else { + dsl_dir_t *dd; + const char *tail; + dsl_dataset_t *origin = NULL; + + VERIFY0(dsl_dir_hold(dp, tofs, FTAG, &dd, &tail)); + + if (drba->drba_origin != NULL) { + VERIFY0(dsl_dataset_hold(dp, drba->drba_origin, + FTAG, &origin)); + } + + /* Create new dataset. */ + dsobj = dsl_dataset_create_sync(dd, + strrchr(tofs, '/') + 1, + origin, crflags, drba->drba_cred, tx); + if (origin != NULL) + dsl_dataset_rele(origin, FTAG); + dsl_dir_rele(dd, FTAG); + drba->drba_cookie->drc_newfs = B_TRUE; + } + VERIFY0(dsl_dataset_own_obj(dp, dsobj, dmu_recv_tag, &newds)); + + if (drba->drba_cookie->drc_resumable) { + dsl_dataset_zapify(newds, tx); + if (drrb->drr_fromguid != 0) { + VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_FROMGUID, + 8, 1, &drrb->drr_fromguid, tx)); + } + VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_TOGUID, + 8, 1, &drrb->drr_toguid, tx)); + VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_TONAME, + 1, strlen(drrb->drr_toname) + 1, drrb->drr_toname, tx)); + uint64_t one = 1; + uint64_t zero = 0; + VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_OBJECT, + 8, 1, &one, tx)); + VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_OFFSET, + 8, 1, &zero, tx)); + VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_BYTES, + 8, 1, &zero, tx)); + if (DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo) & + DMU_BACKUP_FEATURE_LARGE_BLOCKS) { + VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_LARGEBLOCK, + 8, 1, &one, tx)); + } + if (DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo) & + DMU_BACKUP_FEATURE_EMBED_DATA) { + VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_EMBEDOK, + 8, 1, &one, tx)); + } + if (DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo) & + DMU_BACKUP_FEATURE_COMPRESSED) { + VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_COMPRESSOK, + 8, 1, &one, tx)); + } + } + + dmu_buf_will_dirty(newds->ds_dbuf, tx); + dsl_dataset_phys(newds)->ds_flags |= DS_FLAG_INCONSISTENT; + + /* + * If we actually created a non-clone, we need to create the + * objset in our new dataset. + */ + rrw_enter(&newds->ds_bp_rwlock, RW_READER, FTAG); + if (BP_IS_HOLE(dsl_dataset_get_blkptr(newds))) { + (void) dmu_objset_create_impl(dp->dp_spa, + newds, dsl_dataset_get_blkptr(newds), drrb->drr_type, tx); + } + rrw_exit(&newds->ds_bp_rwlock, FTAG); + + drba->drba_cookie->drc_ds = newds; + + spa_history_log_internal_ds(newds, "receive", tx, ""); +} + +static int +dmu_recv_resume_begin_check(void *arg, dmu_tx_t *tx) +{ + dmu_recv_begin_arg_t *drba = arg; + dsl_pool_t *dp = dmu_tx_pool(tx); + struct drr_begin *drrb = drba->drba_cookie->drc_drrb; + int error; + uint64_t featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo); + dsl_dataset_t *ds; + const char *tofs = drba->drba_cookie->drc_tofs; + + /* 6 extra bytes for /%recv */ + char recvname[ZFS_MAX_DATASET_NAME_LEN + 6]; + + /* already checked */ + ASSERT3U(drrb->drr_magic, ==, DMU_BACKUP_MAGIC); + ASSERT(featureflags & DMU_BACKUP_FEATURE_RESUMING); + + if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == + DMU_COMPOUNDSTREAM || + drrb->drr_type >= DMU_OST_NUMTYPES) + return (SET_ERROR(EINVAL)); + + /* Verify pool version supports SA if SA_SPILL feature set */ + if ((featureflags & DMU_BACKUP_FEATURE_SA_SPILL) && + spa_version(dp->dp_spa) < SPA_VERSION_SA) + return (SET_ERROR(ENOTSUP)); + + /* + * The receiving code doesn't know how to translate a WRITE_EMBEDDED + * record to a plain WRITE record, so the pool must have the + * EMBEDDED_DATA feature enabled if the stream has WRITE_EMBEDDED + * records. Same with WRITE_EMBEDDED records that use LZ4 compression. + */ + if ((featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) && + !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EMBEDDED_DATA)) + return (SET_ERROR(ENOTSUP)); + if ((featureflags & DMU_BACKUP_FEATURE_LZ4) && + !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LZ4_COMPRESS)) + return (SET_ERROR(ENOTSUP)); + + /* + * The receiving code doesn't know how to translate large blocks + * to smaller ones, so the pool must have the LARGE_BLOCKS + * feature enabled if the stream has LARGE_BLOCKS. Same with + * large dnodes. + */ + if ((featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) && + !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_BLOCKS)) + return (SET_ERROR(ENOTSUP)); + if ((featureflags & DMU_BACKUP_FEATURE_LARGE_DNODE) && + !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_DNODE)) + return (SET_ERROR(ENOTSUP)); + + (void) snprintf(recvname, sizeof (recvname), "%s/%s", + tofs, recv_clone_name); + + if (dsl_dataset_hold(dp, recvname, FTAG, &ds) != 0) { + /* %recv does not exist; continue in tofs */ + error = dsl_dataset_hold(dp, tofs, FTAG, &ds); + if (error != 0) + return (error); + } + + /* check that ds is marked inconsistent */ + if (!DS_IS_INCONSISTENT(ds)) { + dsl_dataset_rele(ds, FTAG); + return (SET_ERROR(EINVAL)); + } + + /* check that there is resuming data, and that the toguid matches */ + if (!dsl_dataset_is_zapified(ds)) { + dsl_dataset_rele(ds, FTAG); + return (SET_ERROR(EINVAL)); + } + uint64_t val; + error = zap_lookup(dp->dp_meta_objset, ds->ds_object, + DS_FIELD_RESUME_TOGUID, sizeof (val), 1, &val); + if (error != 0 || drrb->drr_toguid != val) { + dsl_dataset_rele(ds, FTAG); + return (SET_ERROR(EINVAL)); + } + + /* + * Check if the receive is still running. If so, it will be owned. + * Note that nothing else can own the dataset (e.g. after the receive + * fails) because it will be marked inconsistent. + */ + if (dsl_dataset_has_owner(ds)) { + dsl_dataset_rele(ds, FTAG); + return (SET_ERROR(EBUSY)); + } + + /* There should not be any snapshots of this fs yet. */ + if (ds->ds_prev != NULL && ds->ds_prev->ds_dir == ds->ds_dir) { + dsl_dataset_rele(ds, FTAG); + return (SET_ERROR(EINVAL)); + } + + /* + * Note: resume point will be checked when we process the first WRITE + * record. + */ + + /* check that the origin matches */ + val = 0; + (void) zap_lookup(dp->dp_meta_objset, ds->ds_object, + DS_FIELD_RESUME_FROMGUID, sizeof (val), 1, &val); + if (drrb->drr_fromguid != val) { + dsl_dataset_rele(ds, FTAG); + return (SET_ERROR(EINVAL)); + } + + dsl_dataset_rele(ds, FTAG); + return (0); +} + +static void +dmu_recv_resume_begin_sync(void *arg, dmu_tx_t *tx) +{ + dmu_recv_begin_arg_t *drba = arg; + dsl_pool_t *dp = dmu_tx_pool(tx); + const char *tofs = drba->drba_cookie->drc_tofs; + dsl_dataset_t *ds; + uint64_t dsobj; + /* 6 extra bytes for /%recv */ + char recvname[ZFS_MAX_DATASET_NAME_LEN + 6]; + + (void) snprintf(recvname, sizeof (recvname), "%s/%s", + tofs, recv_clone_name); + + if (dsl_dataset_hold(dp, recvname, FTAG, &ds) != 0) { + /* %recv does not exist; continue in tofs */ + VERIFY0(dsl_dataset_hold(dp, tofs, FTAG, &ds)); + drba->drba_cookie->drc_newfs = B_TRUE; + } + + /* clear the inconsistent flag so that we can own it */ + ASSERT(DS_IS_INCONSISTENT(ds)); + dmu_buf_will_dirty(ds->ds_dbuf, tx); + dsl_dataset_phys(ds)->ds_flags &= ~DS_FLAG_INCONSISTENT; + dsobj = ds->ds_object; + dsl_dataset_rele(ds, FTAG); + + VERIFY0(dsl_dataset_own_obj(dp, dsobj, dmu_recv_tag, &ds)); + + dmu_buf_will_dirty(ds->ds_dbuf, tx); + dsl_dataset_phys(ds)->ds_flags |= DS_FLAG_INCONSISTENT; + + rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG); + ASSERT(!BP_IS_HOLE(dsl_dataset_get_blkptr(ds))); + rrw_exit(&ds->ds_bp_rwlock, FTAG); + + drba->drba_cookie->drc_ds = ds; + + spa_history_log_internal_ds(ds, "resume receive", tx, ""); +} + +/* + * NB: callers *MUST* call dmu_recv_stream() if dmu_recv_begin() + * succeeds; otherwise we will leak the holds on the datasets. + */ +int +dmu_recv_begin(char *tofs, char *tosnap, dmu_replay_record_t *drr_begin, + boolean_t force, boolean_t resumable, char *origin, dmu_recv_cookie_t *drc) +{ + dmu_recv_begin_arg_t drba = { 0 }; + + bzero(drc, sizeof (dmu_recv_cookie_t)); + drc->drc_drr_begin = drr_begin; + drc->drc_drrb = &drr_begin->drr_u.drr_begin; + drc->drc_tosnap = tosnap; + drc->drc_tofs = tofs; + drc->drc_force = force; + drc->drc_resumable = resumable; + drc->drc_cred = CRED(); + drc->drc_clone = (origin != NULL); + + if (drc->drc_drrb->drr_magic == BSWAP_64(DMU_BACKUP_MAGIC)) { + drc->drc_byteswap = B_TRUE; + (void) fletcher_4_incremental_byteswap(drr_begin, + sizeof (dmu_replay_record_t), &drc->drc_cksum); + byteswap_record(drr_begin); + } else if (drc->drc_drrb->drr_magic == DMU_BACKUP_MAGIC) { + (void) fletcher_4_incremental_native(drr_begin, + sizeof (dmu_replay_record_t), &drc->drc_cksum); + } else { + return (SET_ERROR(EINVAL)); + } + + drba.drba_origin = origin; + drba.drba_cookie = drc; + drba.drba_cred = CRED(); + + if (DMU_GET_FEATUREFLAGS(drc->drc_drrb->drr_versioninfo) & + DMU_BACKUP_FEATURE_RESUMING) { + return (dsl_sync_task(tofs, + dmu_recv_resume_begin_check, dmu_recv_resume_begin_sync, + &drba, 5, ZFS_SPACE_CHECK_NORMAL)); + } else { + return (dsl_sync_task(tofs, + dmu_recv_begin_check, dmu_recv_begin_sync, + &drba, 5, ZFS_SPACE_CHECK_NORMAL)); + } +} + +struct receive_record_arg { + dmu_replay_record_t header; + void *payload; /* Pointer to a buffer containing the payload */ + /* + * If the record is a write, pointer to the arc_buf_t containing the + * payload. + */ + arc_buf_t *write_buf; + int payload_size; + uint64_t bytes_read; /* bytes read from stream when record created */ + boolean_t eos_marker; /* Marks the end of the stream */ + bqueue_node_t node; +}; + +struct receive_writer_arg { + objset_t *os; + boolean_t byteswap; + bqueue_t q; + + /* + * These three args are used to signal to the main thread that we're + * done. + */ + kmutex_t mutex; + kcondvar_t cv; + boolean_t done; + + int err; + /* A map from guid to dataset to help handle dedup'd streams. */ + avl_tree_t *guid_to_ds_map; + boolean_t resumable; + uint64_t last_object; + uint64_t last_offset; + uint64_t max_object; /* highest object ID referenced in stream */ + uint64_t bytes_read; /* bytes read when current record created */ +}; + +struct objlist { + list_t list; /* List of struct receive_objnode. */ + /* + * Last object looked up. Used to assert that objects are being looked + * up in ascending order. + */ + uint64_t last_lookup; +}; + +struct receive_objnode { + list_node_t node; + uint64_t object; +}; + +struct receive_arg { + objset_t *os; + vnode_t *vp; /* The vnode to read the stream from */ + uint64_t voff; /* The current offset in the stream */ + uint64_t bytes_read; + /* + * A record that has had its payload read in, but hasn't yet been handed + * off to the worker thread. + */ + struct receive_record_arg *rrd; + /* A record that has had its header read in, but not its payload. */ + struct receive_record_arg *next_rrd; + zio_cksum_t cksum; + zio_cksum_t prev_cksum; + int err; + boolean_t byteswap; + /* Sorted list of objects not to issue prefetches for. */ + struct objlist ignore_objlist; +}; + +typedef struct guid_map_entry { + uint64_t guid; + dsl_dataset_t *gme_ds; + avl_node_t avlnode; +} guid_map_entry_t; + +static int +guid_compare(const void *arg1, const void *arg2) +{ + const guid_map_entry_t *gmep1 = arg1; + const guid_map_entry_t *gmep2 = arg2; + + if (gmep1->guid < gmep2->guid) + return (-1); + else if (gmep1->guid > gmep2->guid) + return (1); + return (0); +} + +static void +free_guid_map_onexit(void *arg) +{ + avl_tree_t *ca = arg; + void *cookie = NULL; + guid_map_entry_t *gmep; + + while ((gmep = avl_destroy_nodes(ca, &cookie)) != NULL) { + dsl_dataset_long_rele(gmep->gme_ds, gmep); + dsl_dataset_rele(gmep->gme_ds, gmep); + kmem_free(gmep, sizeof (guid_map_entry_t)); + } + avl_destroy(ca); + kmem_free(ca, sizeof (avl_tree_t)); +} + +static int +receive_read(struct receive_arg *ra, int len, void *buf) +{ + int done = 0; + + /* + * The code doesn't rely on this (lengths being multiples of 8). See + * comment in dump_bytes. + */ + ASSERT0(len % 8); + + while (done < len) { + ssize_t resid; + + ra->err = vn_rdwr(UIO_READ, ra->vp, + (char *)buf + done, len - done, + ra->voff, UIO_SYSSPACE, FAPPEND, + RLIM64_INFINITY, CRED(), &resid); + + if (resid == len - done) { + /* + * Note: ECKSUM indicates that the receive + * was interrupted and can potentially be resumed. + */ + ra->err = SET_ERROR(ECKSUM); + } + ra->voff += len - done - resid; + done = len - resid; + if (ra->err != 0) + return (ra->err); + } + + ra->bytes_read += len; + + ASSERT3U(done, ==, len); + return (0); +} + +static void +byteswap_record(dmu_replay_record_t *drr) +{ +#define DO64(X) (drr->drr_u.X = BSWAP_64(drr->drr_u.X)) +#define DO32(X) (drr->drr_u.X = BSWAP_32(drr->drr_u.X)) + drr->drr_type = BSWAP_32(drr->drr_type); + drr->drr_payloadlen = BSWAP_32(drr->drr_payloadlen); + + switch (drr->drr_type) { + case DRR_BEGIN: + DO64(drr_begin.drr_magic); + DO64(drr_begin.drr_versioninfo); + DO64(drr_begin.drr_creation_time); + DO32(drr_begin.drr_type); + DO32(drr_begin.drr_flags); + DO64(drr_begin.drr_toguid); + DO64(drr_begin.drr_fromguid); + break; + case DRR_OBJECT: + DO64(drr_object.drr_object); + DO32(drr_object.drr_type); + DO32(drr_object.drr_bonustype); + DO32(drr_object.drr_blksz); + DO32(drr_object.drr_bonuslen); + DO64(drr_object.drr_toguid); + break; + case DRR_FREEOBJECTS: + DO64(drr_freeobjects.drr_firstobj); + DO64(drr_freeobjects.drr_numobjs); + DO64(drr_freeobjects.drr_toguid); + break; + case DRR_WRITE: + DO64(drr_write.drr_object); + DO32(drr_write.drr_type); + DO64(drr_write.drr_offset); + DO64(drr_write.drr_logical_size); + DO64(drr_write.drr_toguid); + ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_write.drr_key.ddk_cksum); + DO64(drr_write.drr_key.ddk_prop); + DO64(drr_write.drr_compressed_size); + break; + case DRR_WRITE_BYREF: + DO64(drr_write_byref.drr_object); + DO64(drr_write_byref.drr_offset); + DO64(drr_write_byref.drr_length); + DO64(drr_write_byref.drr_toguid); + DO64(drr_write_byref.drr_refguid); + DO64(drr_write_byref.drr_refobject); + DO64(drr_write_byref.drr_refoffset); + ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_write_byref. + drr_key.ddk_cksum); + DO64(drr_write_byref.drr_key.ddk_prop); + break; + case DRR_WRITE_EMBEDDED: + DO64(drr_write_embedded.drr_object); + DO64(drr_write_embedded.drr_offset); + DO64(drr_write_embedded.drr_length); + DO64(drr_write_embedded.drr_toguid); + DO32(drr_write_embedded.drr_lsize); + DO32(drr_write_embedded.drr_psize); + break; + case DRR_FREE: + DO64(drr_free.drr_object); + DO64(drr_free.drr_offset); + DO64(drr_free.drr_length); + DO64(drr_free.drr_toguid); + break; + case DRR_SPILL: + DO64(drr_spill.drr_object); + DO64(drr_spill.drr_length); + DO64(drr_spill.drr_toguid); + break; + case DRR_END: + DO64(drr_end.drr_toguid); + ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_end.drr_checksum); + break; + } + + if (drr->drr_type != DRR_BEGIN) { + ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_checksum.drr_checksum); + } + +#undef DO64 +#undef DO32 +} + +static inline uint8_t +deduce_nblkptr(dmu_object_type_t bonus_type, uint64_t bonus_size) +{ + if (bonus_type == DMU_OT_SA) { + return (1); + } else { + return (1 + + ((DN_OLD_MAX_BONUSLEN - + MIN(DN_OLD_MAX_BONUSLEN, bonus_size)) >> SPA_BLKPTRSHIFT)); + } +} + +static void +save_resume_state(struct receive_writer_arg *rwa, + uint64_t object, uint64_t offset, dmu_tx_t *tx) +{ + int txgoff = dmu_tx_get_txg(tx) & TXG_MASK; + + if (!rwa->resumable) + return; + + /* + * We use ds_resume_bytes[] != 0 to indicate that we need to + * update this on disk, so it must not be 0. + */ + ASSERT(rwa->bytes_read != 0); + + /* + * We only resume from write records, which have a valid + * (non-meta-dnode) object number. + */ + ASSERT(object != 0); + + /* + * For resuming to work correctly, we must receive records in order, + * sorted by object,offset. This is checked by the callers, but + * assert it here for good measure. + */ + ASSERT3U(object, >=, rwa->os->os_dsl_dataset->ds_resume_object[txgoff]); + ASSERT(object != rwa->os->os_dsl_dataset->ds_resume_object[txgoff] || + offset >= rwa->os->os_dsl_dataset->ds_resume_offset[txgoff]); + ASSERT3U(rwa->bytes_read, >=, + rwa->os->os_dsl_dataset->ds_resume_bytes[txgoff]); + + rwa->os->os_dsl_dataset->ds_resume_object[txgoff] = object; + rwa->os->os_dsl_dataset->ds_resume_offset[txgoff] = offset; + rwa->os->os_dsl_dataset->ds_resume_bytes[txgoff] = rwa->bytes_read; +} + +static int +receive_object(struct receive_writer_arg *rwa, struct drr_object *drro, + void *data) +{ + dmu_object_info_t doi; + dmu_tx_t *tx; + uint64_t object; + int err; + uint8_t dn_slots = drro->drr_dn_slots != 0 ? + drro->drr_dn_slots : DNODE_MIN_SLOTS; + + if (drro->drr_type == DMU_OT_NONE || + !DMU_OT_IS_VALID(drro->drr_type) || + !DMU_OT_IS_VALID(drro->drr_bonustype) || + drro->drr_checksumtype >= ZIO_CHECKSUM_FUNCTIONS || + drro->drr_compress >= ZIO_COMPRESS_FUNCTIONS || + P2PHASE(drro->drr_blksz, SPA_MINBLOCKSIZE) || + drro->drr_blksz < SPA_MINBLOCKSIZE || + drro->drr_blksz > spa_maxblocksize(dmu_objset_spa(rwa->os)) || + drro->drr_bonuslen > + DN_BONUS_SIZE(spa_maxdnodesize(dmu_objset_spa(rwa->os))) || + dn_slots > + (spa_maxdnodesize(dmu_objset_spa(rwa->os)) >> DNODE_SHIFT)) { + return (SET_ERROR(EINVAL)); + } + + err = dmu_object_info(rwa->os, drro->drr_object, &doi); + + if (err != 0 && err != ENOENT && err != EEXIST) + return (SET_ERROR(EINVAL)); + + if (drro->drr_object > rwa->max_object) + rwa->max_object = drro->drr_object; + + /* + * If we are losing blkptrs or changing the block size this must + * be a new file instance. We must clear out the previous file + * contents before we can change this type of metadata in the dnode. + */ + if (err == 0) { + int nblkptr; + + object = drro->drr_object; + + nblkptr = deduce_nblkptr(drro->drr_bonustype, + drro->drr_bonuslen); + + if (drro->drr_blksz != doi.doi_data_block_size || + nblkptr < doi.doi_nblkptr || + dn_slots != doi.doi_dnodesize >> DNODE_SHIFT) { + err = dmu_free_long_range(rwa->os, drro->drr_object, + 0, DMU_OBJECT_END); + if (err != 0) + return (SET_ERROR(EINVAL)); + } + } else if (err == EEXIST) { + /* + * The object requested is currently an interior slot of a + * multi-slot dnode. This will be resolved when the next txg + * is synced out, since the send stream will have told us + * to free this slot when we freed the associated dnode + * earlier in the stream. + */ + txg_wait_synced(dmu_objset_pool(rwa->os), 0); + object = drro->drr_object; + } else { + /* object is free and we are about to allocate a new one */ + object = DMU_NEW_OBJECT; + } + + /* + * If this is a multi-slot dnode there is a chance that this + * object will expand into a slot that is already used by + * another object from the previous snapshot. We must free + * these objects before we attempt to allocate the new dnode. + */ + if (dn_slots > 1) { + boolean_t need_sync = B_FALSE; + + for (uint64_t slot = drro->drr_object + 1; + slot < drro->drr_object + dn_slots; + slot++) { + dmu_object_info_t slot_doi; + + err = dmu_object_info(rwa->os, slot, &slot_doi); + if (err == ENOENT || err == EEXIST) + continue; + else if (err != 0) + return (err); + + err = dmu_free_long_object(rwa->os, slot); + + if (err != 0) + return (err); + + need_sync = B_TRUE; + } + + if (need_sync) + txg_wait_synced(dmu_objset_pool(rwa->os), 0); + } + + tx = dmu_tx_create(rwa->os); + dmu_tx_hold_bonus(tx, object); + err = dmu_tx_assign(tx, TXG_WAIT); + if (err != 0) { + dmu_tx_abort(tx); + return (err); + } + + if (object == DMU_NEW_OBJECT) { + /* currently free, want to be allocated */ + err = dmu_object_claim_dnsize(rwa->os, drro->drr_object, + drro->drr_type, drro->drr_blksz, + drro->drr_bonustype, drro->drr_bonuslen, + dn_slots << DNODE_SHIFT, tx); + } else if (drro->drr_type != doi.doi_type || + drro->drr_blksz != doi.doi_data_block_size || + drro->drr_bonustype != doi.doi_bonus_type || + drro->drr_bonuslen != doi.doi_bonus_size || + drro->drr_dn_slots != (doi.doi_dnodesize >> DNODE_SHIFT)) { + /* currently allocated, but with different properties */ + err = dmu_object_reclaim_dnsize(rwa->os, drro->drr_object, + drro->drr_type, drro->drr_blksz, + drro->drr_bonustype, drro->drr_bonuslen, + drro->drr_dn_slots << DNODE_SHIFT, tx); + } + if (err != 0) { + dmu_tx_commit(tx); + return (SET_ERROR(EINVAL)); + } + + dmu_object_set_checksum(rwa->os, drro->drr_object, + drro->drr_checksumtype, tx); + dmu_object_set_compress(rwa->os, drro->drr_object, + drro->drr_compress, tx); + + if (data != NULL) { + dmu_buf_t *db; + + VERIFY0(dmu_bonus_hold(rwa->os, drro->drr_object, FTAG, &db)); + dmu_buf_will_dirty(db, tx); + + ASSERT3U(db->db_size, >=, drro->drr_bonuslen); + bcopy(data, db->db_data, drro->drr_bonuslen); + if (rwa->byteswap) { + dmu_object_byteswap_t byteswap = + DMU_OT_BYTESWAP(drro->drr_bonustype); + dmu_ot_byteswap[byteswap].ob_func(db->db_data, + drro->drr_bonuslen); + } + dmu_buf_rele(db, FTAG); + } + dmu_tx_commit(tx); + + return (0); +} + +/* ARGSUSED */ +static int +receive_freeobjects(struct receive_writer_arg *rwa, + struct drr_freeobjects *drrfo) +{ + uint64_t obj; + int next_err = 0; + + if (drrfo->drr_firstobj + drrfo->drr_numobjs < drrfo->drr_firstobj) + return (SET_ERROR(EINVAL)); + + for (obj = drrfo->drr_firstobj == 0 ? 1 : drrfo->drr_firstobj; + obj < drrfo->drr_firstobj + drrfo->drr_numobjs && next_err == 0; + next_err = dmu_object_next(rwa->os, &obj, FALSE, 0)) { + int err; + + err = dmu_object_info(rwa->os, obj, NULL); + if (err == ENOENT) + continue; + else if (err != 0) + return (err); + + err = dmu_free_long_object(rwa->os, obj); + if (err != 0) + return (err); + + if (obj > rwa->max_object) + rwa->max_object = obj; + } + if (next_err != ESRCH) + return (next_err); + return (0); +} + +static int +receive_write(struct receive_writer_arg *rwa, struct drr_write *drrw, + arc_buf_t *abuf) +{ + dmu_tx_t *tx; + int err; + + if (drrw->drr_offset + drrw->drr_logical_size < drrw->drr_offset || + !DMU_OT_IS_VALID(drrw->drr_type)) + return (SET_ERROR(EINVAL)); + + /* + * For resuming to work, records must be in increasing order + * by (object, offset). + */ + if (drrw->drr_object < rwa->last_object || + (drrw->drr_object == rwa->last_object && + drrw->drr_offset < rwa->last_offset)) { + return (SET_ERROR(EINVAL)); + } + rwa->last_object = drrw->drr_object; + rwa->last_offset = drrw->drr_offset; + + if (rwa->last_object > rwa->max_object) + rwa->max_object = rwa->last_object; + + if (dmu_object_info(rwa->os, drrw->drr_object, NULL) != 0) + return (SET_ERROR(EINVAL)); + + tx = dmu_tx_create(rwa->os); + + dmu_tx_hold_write(tx, drrw->drr_object, + drrw->drr_offset, drrw->drr_logical_size); + err = dmu_tx_assign(tx, TXG_WAIT); + if (err != 0) { + dmu_tx_abort(tx); + return (err); + } + if (rwa->byteswap) { + dmu_object_byteswap_t byteswap = + DMU_OT_BYTESWAP(drrw->drr_type); + dmu_ot_byteswap[byteswap].ob_func(abuf->b_data, + DRR_WRITE_PAYLOAD_SIZE(drrw)); + } + + /* use the bonus buf to look up the dnode in dmu_assign_arcbuf */ + dmu_buf_t *bonus; + if (dmu_bonus_hold(rwa->os, drrw->drr_object, FTAG, &bonus) != 0) + return (SET_ERROR(EINVAL)); + dmu_assign_arcbuf(bonus, drrw->drr_offset, abuf, tx); + + /* + * Note: If the receive fails, we want the resume stream to start + * with the same record that we last successfully received (as opposed + * to the next record), so that we can verify that we are + * resuming from the correct location. + */ + save_resume_state(rwa, drrw->drr_object, drrw->drr_offset, tx); + dmu_tx_commit(tx); + dmu_buf_rele(bonus, FTAG); + + return (0); +} + +/* + * Handle a DRR_WRITE_BYREF record. This record is used in dedup'ed + * streams to refer to a copy of the data that is already on the + * system because it came in earlier in the stream. This function + * finds the earlier copy of the data, and uses that copy instead of + * data from the stream to fulfill this write. + */ +static int +receive_write_byref(struct receive_writer_arg *rwa, + struct drr_write_byref *drrwbr) +{ + dmu_tx_t *tx; + int err; + guid_map_entry_t gmesrch; + guid_map_entry_t *gmep; + avl_index_t where; + objset_t *ref_os = NULL; + dmu_buf_t *dbp; + + if (drrwbr->drr_offset + drrwbr->drr_length < drrwbr->drr_offset) + return (SET_ERROR(EINVAL)); + + /* + * If the GUID of the referenced dataset is different from the + * GUID of the target dataset, find the referenced dataset. + */ + if (drrwbr->drr_toguid != drrwbr->drr_refguid) { + gmesrch.guid = drrwbr->drr_refguid; + if ((gmep = avl_find(rwa->guid_to_ds_map, &gmesrch, + &where)) == NULL) { + return (SET_ERROR(EINVAL)); + } + if (dmu_objset_from_ds(gmep->gme_ds, &ref_os)) + return (SET_ERROR(EINVAL)); + } else { + ref_os = rwa->os; + } + + if (drrwbr->drr_object > rwa->max_object) + rwa->max_object = drrwbr->drr_object; + + err = dmu_buf_hold(ref_os, drrwbr->drr_refobject, + drrwbr->drr_refoffset, FTAG, &dbp, DMU_READ_PREFETCH); + if (err != 0) + return (err); + + tx = dmu_tx_create(rwa->os); + + dmu_tx_hold_write(tx, drrwbr->drr_object, + drrwbr->drr_offset, drrwbr->drr_length); + err = dmu_tx_assign(tx, TXG_WAIT); + if (err != 0) { + dmu_tx_abort(tx); + return (err); + } + dmu_write(rwa->os, drrwbr->drr_object, + drrwbr->drr_offset, drrwbr->drr_length, dbp->db_data, tx); + dmu_buf_rele(dbp, FTAG); + + /* See comment in restore_write. */ + save_resume_state(rwa, drrwbr->drr_object, drrwbr->drr_offset, tx); + dmu_tx_commit(tx); + return (0); +} + +static int +receive_write_embedded(struct receive_writer_arg *rwa, + struct drr_write_embedded *drrwe, void *data) +{ + dmu_tx_t *tx; + int err; + + if (drrwe->drr_offset + drrwe->drr_length < drrwe->drr_offset) + return (EINVAL); + + if (drrwe->drr_psize > BPE_PAYLOAD_SIZE) + return (EINVAL); + + if (drrwe->drr_etype >= NUM_BP_EMBEDDED_TYPES) + return (EINVAL); + if (drrwe->drr_compression >= ZIO_COMPRESS_FUNCTIONS) + return (EINVAL); + + if (drrwe->drr_object > rwa->max_object) + rwa->max_object = drrwe->drr_object; + + tx = dmu_tx_create(rwa->os); + + dmu_tx_hold_write(tx, drrwe->drr_object, + drrwe->drr_offset, drrwe->drr_length); + err = dmu_tx_assign(tx, TXG_WAIT); + if (err != 0) { + dmu_tx_abort(tx); + return (err); + } + + dmu_write_embedded(rwa->os, drrwe->drr_object, + drrwe->drr_offset, data, drrwe->drr_etype, + drrwe->drr_compression, drrwe->drr_lsize, drrwe->drr_psize, + rwa->byteswap ^ ZFS_HOST_BYTEORDER, tx); + + /* See comment in restore_write. */ + save_resume_state(rwa, drrwe->drr_object, drrwe->drr_offset, tx); + dmu_tx_commit(tx); + return (0); +} + +static int +receive_spill(struct receive_writer_arg *rwa, struct drr_spill *drrs, + void *data) +{ + dmu_tx_t *tx; + dmu_buf_t *db, *db_spill; + int err; + + if (drrs->drr_length < SPA_MINBLOCKSIZE || + drrs->drr_length > spa_maxblocksize(dmu_objset_spa(rwa->os))) + return (SET_ERROR(EINVAL)); + + if (dmu_object_info(rwa->os, drrs->drr_object, NULL) != 0) + return (SET_ERROR(EINVAL)); + + if (drrs->drr_object > rwa->max_object) + rwa->max_object = drrs->drr_object; + + VERIFY0(dmu_bonus_hold(rwa->os, drrs->drr_object, FTAG, &db)); + if ((err = dmu_spill_hold_by_bonus(db, FTAG, &db_spill)) != 0) { + dmu_buf_rele(db, FTAG); + return (err); + } + + tx = dmu_tx_create(rwa->os); + + dmu_tx_hold_spill(tx, db->db_object); + + err = dmu_tx_assign(tx, TXG_WAIT); + if (err != 0) { + dmu_buf_rele(db, FTAG); + dmu_buf_rele(db_spill, FTAG); + dmu_tx_abort(tx); + return (err); + } + dmu_buf_will_dirty(db_spill, tx); + + if (db_spill->db_size < drrs->drr_length) + VERIFY(0 == dbuf_spill_set_blksz(db_spill, + drrs->drr_length, tx)); + bcopy(data, db_spill->db_data, drrs->drr_length); + + dmu_buf_rele(db, FTAG); + dmu_buf_rele(db_spill, FTAG); + + dmu_tx_commit(tx); + return (0); +} + +/* ARGSUSED */ +static int +receive_free(struct receive_writer_arg *rwa, struct drr_free *drrf) +{ + int err; + + if (drrf->drr_length != -1ULL && + drrf->drr_offset + drrf->drr_length < drrf->drr_offset) + return (SET_ERROR(EINVAL)); + + if (dmu_object_info(rwa->os, drrf->drr_object, NULL) != 0) + return (SET_ERROR(EINVAL)); + + if (drrf->drr_object > rwa->max_object) + rwa->max_object = drrf->drr_object; + + err = dmu_free_long_range(rwa->os, drrf->drr_object, + drrf->drr_offset, drrf->drr_length); + + return (err); +} + +/* used to destroy the drc_ds on error */ +static void +dmu_recv_cleanup_ds(dmu_recv_cookie_t *drc) +{ + if (drc->drc_resumable) { + /* wait for our resume state to be written to disk */ + txg_wait_synced(drc->drc_ds->ds_dir->dd_pool, 0); + dsl_dataset_disown(drc->drc_ds, dmu_recv_tag); + } else { + char name[ZFS_MAX_DATASET_NAME_LEN]; + dsl_dataset_name(drc->drc_ds, name); + dsl_dataset_disown(drc->drc_ds, dmu_recv_tag); + (void) dsl_destroy_head(name); + } +} + +static void +receive_cksum(struct receive_arg *ra, int len, void *buf) +{ + if (ra->byteswap) { + (void) fletcher_4_incremental_byteswap(buf, len, &ra->cksum); + } else { + (void) fletcher_4_incremental_native(buf, len, &ra->cksum); + } +} + +/* + * Read the payload into a buffer of size len, and update the current record's + * payload field. + * Allocate ra->next_rrd and read the next record's header into + * ra->next_rrd->header. + * Verify checksum of payload and next record. + */ +static int +receive_read_payload_and_next_header(struct receive_arg *ra, int len, void *buf) +{ + int err; + + if (len != 0) { + ASSERT3U(len, <=, SPA_MAXBLOCKSIZE); + err = receive_read(ra, len, buf); + if (err != 0) + return (err); + receive_cksum(ra, len, buf); + + /* note: rrd is NULL when reading the begin record's payload */ + if (ra->rrd != NULL) { + ra->rrd->payload = buf; + ra->rrd->payload_size = len; + ra->rrd->bytes_read = ra->bytes_read; + } + } + + ra->prev_cksum = ra->cksum; + + ra->next_rrd = kmem_zalloc(sizeof (*ra->next_rrd), KM_SLEEP); + err = receive_read(ra, sizeof (ra->next_rrd->header), + &ra->next_rrd->header); + ra->next_rrd->bytes_read = ra->bytes_read; + if (err != 0) { + kmem_free(ra->next_rrd, sizeof (*ra->next_rrd)); + ra->next_rrd = NULL; + return (err); + } + if (ra->next_rrd->header.drr_type == DRR_BEGIN) { + kmem_free(ra->next_rrd, sizeof (*ra->next_rrd)); + ra->next_rrd = NULL; + return (SET_ERROR(EINVAL)); + } + + /* + * Note: checksum is of everything up to but not including the + * checksum itself. + */ + ASSERT3U(offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), + ==, sizeof (dmu_replay_record_t) - sizeof (zio_cksum_t)); + receive_cksum(ra, + offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), + &ra->next_rrd->header); + + zio_cksum_t cksum_orig = + ra->next_rrd->header.drr_u.drr_checksum.drr_checksum; + zio_cksum_t *cksump = + &ra->next_rrd->header.drr_u.drr_checksum.drr_checksum; + + if (ra->byteswap) + byteswap_record(&ra->next_rrd->header); + + if ((!ZIO_CHECKSUM_IS_ZERO(cksump)) && + !ZIO_CHECKSUM_EQUAL(ra->cksum, *cksump)) { + kmem_free(ra->next_rrd, sizeof (*ra->next_rrd)); + ra->next_rrd = NULL; + return (SET_ERROR(ECKSUM)); + } + + receive_cksum(ra, sizeof (cksum_orig), &cksum_orig); + + return (0); +} + +static void +objlist_create(struct objlist *list) +{ + list_create(&list->list, sizeof (struct receive_objnode), + offsetof(struct receive_objnode, node)); + list->last_lookup = 0; +} + +static void +objlist_destroy(struct objlist *list) +{ + for (struct receive_objnode *n = list_remove_head(&list->list); + n != NULL; n = list_remove_head(&list->list)) { + kmem_free(n, sizeof (*n)); + } + list_destroy(&list->list); +} + +/* + * This function looks through the objlist to see if the specified object number + * is contained in the objlist. In the process, it will remove all object + * numbers in the list that are smaller than the specified object number. Thus, + * any lookup of an object number smaller than a previously looked up object + * number will always return false; therefore, all lookups should be done in + * ascending order. + */ +static boolean_t +objlist_exists(struct objlist *list, uint64_t object) +{ + struct receive_objnode *node = list_head(&list->list); + ASSERT3U(object, >=, list->last_lookup); + list->last_lookup = object; + while (node != NULL && node->object < object) { + VERIFY3P(node, ==, list_remove_head(&list->list)); + kmem_free(node, sizeof (*node)); + node = list_head(&list->list); + } + return (node != NULL && node->object == object); +} + +/* + * The objlist is a list of object numbers stored in ascending order. However, + * the insertion of new object numbers does not seek out the correct location to + * store a new object number; instead, it appends it to the list for simplicity. + * Thus, any users must take care to only insert new object numbers in ascending + * order. + */ +static void +objlist_insert(struct objlist *list, uint64_t object) +{ + struct receive_objnode *node = kmem_zalloc(sizeof (*node), KM_SLEEP); + node->object = object; +#ifdef ZFS_DEBUG + struct receive_objnode *last_object = list_tail(&list->list); + uint64_t last_objnum = (last_object != NULL ? last_object->object : 0); + ASSERT3U(node->object, >, last_objnum); +#endif + list_insert_tail(&list->list, node); +} + +/* + * Issue the prefetch reads for any necessary indirect blocks. + * + * We use the object ignore list to tell us whether or not to issue prefetches + * for a given object. We do this for both correctness (in case the blocksize + * of an object has changed) and performance (if the object doesn't exist, don't + * needlessly try to issue prefetches). We also trim the list as we go through + * the stream to prevent it from growing to an unbounded size. + * + * The object numbers within will always be in sorted order, and any write + * records we see will also be in sorted order, but they're not sorted with + * respect to each other (i.e. we can get several object records before + * receiving each object's write records). As a result, once we've reached a + * given object number, we can safely remove any reference to lower object + * numbers in the ignore list. In practice, we receive up to 32 object records + * before receiving write records, so the list can have up to 32 nodes in it. + */ +/* ARGSUSED */ +static void +receive_read_prefetch(struct receive_arg *ra, + uint64_t object, uint64_t offset, uint64_t length) +{ + if (!objlist_exists(&ra->ignore_objlist, object)) { + dmu_prefetch(ra->os, object, 1, offset, length, + ZIO_PRIORITY_SYNC_READ); + } +} + +/* + * Read records off the stream, issuing any necessary prefetches. + */ +static int +receive_read_record(struct receive_arg *ra) +{ + int err; + + switch (ra->rrd->header.drr_type) { + case DRR_OBJECT: + { + struct drr_object *drro = &ra->rrd->header.drr_u.drr_object; + uint32_t size = P2ROUNDUP(drro->drr_bonuslen, 8); + void *buf = NULL; + dmu_object_info_t doi; + + if (size > 0) + buf = kmem_zalloc(size, KM_SLEEP); + + err = receive_read_payload_and_next_header(ra, size, buf); + if (err != 0) { + kmem_free(buf, size); + return (err); + } + err = dmu_object_info(ra->os, drro->drr_object, &doi); + /* + * See receive_read_prefetch for an explanation why we're + * storing this object in the ignore_obj_list. + */ + if (err == ENOENT || + (err == 0 && doi.doi_data_block_size != drro->drr_blksz)) { + objlist_insert(&ra->ignore_objlist, drro->drr_object); + err = 0; + } + return (err); + } + case DRR_FREEOBJECTS: + { + err = receive_read_payload_and_next_header(ra, 0, NULL); + return (err); + } + case DRR_WRITE: + { + struct drr_write *drrw = &ra->rrd->header.drr_u.drr_write; + arc_buf_t *abuf; + boolean_t is_meta = DMU_OT_IS_METADATA(drrw->drr_type); + if (DRR_WRITE_COMPRESSED(drrw)) { + ASSERT3U(drrw->drr_compressed_size, >, 0); + ASSERT3U(drrw->drr_logical_size, >=, + drrw->drr_compressed_size); + ASSERT(!is_meta); + abuf = arc_loan_compressed_buf( + dmu_objset_spa(ra->os), + drrw->drr_compressed_size, drrw->drr_logical_size, + drrw->drr_compressiontype); + } else { + abuf = arc_loan_buf(dmu_objset_spa(ra->os), + is_meta, drrw->drr_logical_size); + } + + err = receive_read_payload_and_next_header(ra, + DRR_WRITE_PAYLOAD_SIZE(drrw), abuf->b_data); + if (err != 0) { + dmu_return_arcbuf(abuf); + return (err); + } + ra->rrd->write_buf = abuf; + receive_read_prefetch(ra, drrw->drr_object, drrw->drr_offset, + drrw->drr_logical_size); + return (err); + } + case DRR_WRITE_BYREF: + { + struct drr_write_byref *drrwb = + &ra->rrd->header.drr_u.drr_write_byref; + err = receive_read_payload_and_next_header(ra, 0, NULL); + receive_read_prefetch(ra, drrwb->drr_object, drrwb->drr_offset, + drrwb->drr_length); + return (err); + } + case DRR_WRITE_EMBEDDED: + { + struct drr_write_embedded *drrwe = + &ra->rrd->header.drr_u.drr_write_embedded; + uint32_t size = P2ROUNDUP(drrwe->drr_psize, 8); + void *buf = kmem_zalloc(size, KM_SLEEP); + + err = receive_read_payload_and_next_header(ra, size, buf); + if (err != 0) { + kmem_free(buf, size); + return (err); + } + + receive_read_prefetch(ra, drrwe->drr_object, drrwe->drr_offset, + drrwe->drr_length); + return (err); + } + case DRR_FREE: + { + /* + * It might be beneficial to prefetch indirect blocks here, but + * we don't really have the data to decide for sure. + */ + err = receive_read_payload_and_next_header(ra, 0, NULL); + return (err); + } + case DRR_END: + { + struct drr_end *drre = &ra->rrd->header.drr_u.drr_end; + if (!ZIO_CHECKSUM_EQUAL(ra->prev_cksum, drre->drr_checksum)) + return (SET_ERROR(ECKSUM)); + return (0); + } + case DRR_SPILL: + { + struct drr_spill *drrs = &ra->rrd->header.drr_u.drr_spill; + void *buf = kmem_zalloc(drrs->drr_length, KM_SLEEP); + err = receive_read_payload_and_next_header(ra, drrs->drr_length, + buf); + if (err != 0) + kmem_free(buf, drrs->drr_length); + return (err); + } + default: + return (SET_ERROR(EINVAL)); + } +} + +/* + * Commit the records to the pool. + */ +static int +receive_process_record(struct receive_writer_arg *rwa, + struct receive_record_arg *rrd) +{ + int err; + + /* Processing in order, therefore bytes_read should be increasing. */ + ASSERT3U(rrd->bytes_read, >=, rwa->bytes_read); + rwa->bytes_read = rrd->bytes_read; + + switch (rrd->header.drr_type) { + case DRR_OBJECT: + { + struct drr_object *drro = &rrd->header.drr_u.drr_object; + err = receive_object(rwa, drro, rrd->payload); + kmem_free(rrd->payload, rrd->payload_size); + rrd->payload = NULL; + return (err); + } + case DRR_FREEOBJECTS: + { + struct drr_freeobjects *drrfo = + &rrd->header.drr_u.drr_freeobjects; + return (receive_freeobjects(rwa, drrfo)); + } + case DRR_WRITE: + { + struct drr_write *drrw = &rrd->header.drr_u.drr_write; + err = receive_write(rwa, drrw, rrd->write_buf); + /* if receive_write() is successful, it consumes the arc_buf */ + if (err != 0) + dmu_return_arcbuf(rrd->write_buf); + rrd->write_buf = NULL; + rrd->payload = NULL; + return (err); + } + case DRR_WRITE_BYREF: + { + struct drr_write_byref *drrwbr = + &rrd->header.drr_u.drr_write_byref; + return (receive_write_byref(rwa, drrwbr)); + } + case DRR_WRITE_EMBEDDED: + { + struct drr_write_embedded *drrwe = + &rrd->header.drr_u.drr_write_embedded; + err = receive_write_embedded(rwa, drrwe, rrd->payload); + kmem_free(rrd->payload, rrd->payload_size); + rrd->payload = NULL; + return (err); + } + case DRR_FREE: + { + struct drr_free *drrf = &rrd->header.drr_u.drr_free; + return (receive_free(rwa, drrf)); + } + case DRR_SPILL: + { + struct drr_spill *drrs = &rrd->header.drr_u.drr_spill; + err = receive_spill(rwa, drrs, rrd->payload); + kmem_free(rrd->payload, rrd->payload_size); + rrd->payload = NULL; + return (err); + } + default: + return (SET_ERROR(EINVAL)); + } +} + +/* + * dmu_recv_stream's worker thread; pull records off the queue, and then call + * receive_process_record When we're done, signal the main thread and exit. + */ +static void +receive_writer_thread(void *arg) +{ + struct receive_writer_arg *rwa = arg; + struct receive_record_arg *rrd; + for (rrd = bqueue_dequeue(&rwa->q); !rrd->eos_marker; + rrd = bqueue_dequeue(&rwa->q)) { + /* + * If there's an error, the main thread will stop putting things + * on the queue, but we need to clear everything in it before we + * can exit. + */ + if (rwa->err == 0) { + rwa->err = receive_process_record(rwa, rrd); + } else if (rrd->write_buf != NULL) { + dmu_return_arcbuf(rrd->write_buf); + rrd->write_buf = NULL; + rrd->payload = NULL; + } else if (rrd->payload != NULL) { + kmem_free(rrd->payload, rrd->payload_size); + rrd->payload = NULL; + } + kmem_free(rrd, sizeof (*rrd)); + } + kmem_free(rrd, sizeof (*rrd)); + mutex_enter(&rwa->mutex); + rwa->done = B_TRUE; + cv_signal(&rwa->cv); + mutex_exit(&rwa->mutex); + thread_exit(); +} + +static int +resume_check(struct receive_arg *ra, nvlist_t *begin_nvl) +{ + uint64_t val; + objset_t *mos = dmu_objset_pool(ra->os)->dp_meta_objset; + uint64_t dsobj = dmu_objset_id(ra->os); + uint64_t resume_obj, resume_off; + + if (nvlist_lookup_uint64(begin_nvl, + "resume_object", &resume_obj) != 0 || + nvlist_lookup_uint64(begin_nvl, + "resume_offset", &resume_off) != 0) { + return (SET_ERROR(EINVAL)); + } + VERIFY0(zap_lookup(mos, dsobj, + DS_FIELD_RESUME_OBJECT, sizeof (val), 1, &val)); + if (resume_obj != val) + return (SET_ERROR(EINVAL)); + VERIFY0(zap_lookup(mos, dsobj, + DS_FIELD_RESUME_OFFSET, sizeof (val), 1, &val)); + if (resume_off != val) + return (SET_ERROR(EINVAL)); + + return (0); +} + +/* + * Read in the stream's records, one by one, and apply them to the pool. There + * are two threads involved; the thread that calls this function will spin up a + * worker thread, read the records off the stream one by one, and issue + * prefetches for any necessary indirect blocks. It will then push the records + * onto an internal blocking queue. The worker thread will pull the records off + * the queue, and actually write the data into the DMU. This way, the worker + * thread doesn't have to wait for reads to complete, since everything it needs + * (the indirect blocks) will be prefetched. + * + * NB: callers *must* call dmu_recv_end() if this succeeds. + */ +int +dmu_recv_stream(dmu_recv_cookie_t *drc, vnode_t *vp, offset_t *voffp, + int cleanup_fd, uint64_t *action_handlep) +{ + int err = 0; + struct receive_arg ra = { 0 }; + struct receive_writer_arg rwa = { 0 }; + int featureflags; + nvlist_t *begin_nvl = NULL; + + ra.byteswap = drc->drc_byteswap; + ra.cksum = drc->drc_cksum; + ra.vp = vp; + ra.voff = *voffp; + + if (dsl_dataset_is_zapified(drc->drc_ds)) { + (void) zap_lookup(drc->drc_ds->ds_dir->dd_pool->dp_meta_objset, + drc->drc_ds->ds_object, DS_FIELD_RESUME_BYTES, + sizeof (ra.bytes_read), 1, &ra.bytes_read); + } + + objlist_create(&ra.ignore_objlist); + + /* these were verified in dmu_recv_begin */ + ASSERT3U(DMU_GET_STREAM_HDRTYPE(drc->drc_drrb->drr_versioninfo), ==, + DMU_SUBSTREAM); + ASSERT3U(drc->drc_drrb->drr_type, <, DMU_OST_NUMTYPES); + + /* + * Open the objset we are modifying. + */ + VERIFY0(dmu_objset_from_ds(drc->drc_ds, &ra.os)); + + ASSERT(dsl_dataset_phys(drc->drc_ds)->ds_flags & DS_FLAG_INCONSISTENT); + + featureflags = DMU_GET_FEATUREFLAGS(drc->drc_drrb->drr_versioninfo); + + /* if this stream is dedup'ed, set up the avl tree for guid mapping */ + if (featureflags & DMU_BACKUP_FEATURE_DEDUP) { + minor_t minor; + + if (cleanup_fd == -1) { + ra.err = SET_ERROR(EBADF); + goto out; + } + ra.err = zfs_onexit_fd_hold(cleanup_fd, &minor); + if (ra.err != 0) { + cleanup_fd = -1; + goto out; + } + + if (*action_handlep == 0) { + rwa.guid_to_ds_map = + kmem_alloc(sizeof (avl_tree_t), KM_SLEEP); + avl_create(rwa.guid_to_ds_map, guid_compare, + sizeof (guid_map_entry_t), + offsetof(guid_map_entry_t, avlnode)); + err = zfs_onexit_add_cb(minor, + free_guid_map_onexit, rwa.guid_to_ds_map, + action_handlep); + if (ra.err != 0) + goto out; + } else { + err = zfs_onexit_cb_data(minor, *action_handlep, + (void **)&rwa.guid_to_ds_map); + if (ra.err != 0) + goto out; + } + + drc->drc_guid_to_ds_map = rwa.guid_to_ds_map; + } + + uint32_t payloadlen = drc->drc_drr_begin->drr_payloadlen; + void *payload = NULL; + if (payloadlen != 0) + payload = kmem_alloc(payloadlen, KM_SLEEP); + + err = receive_read_payload_and_next_header(&ra, payloadlen, payload); + if (err != 0) { + if (payloadlen != 0) + kmem_free(payload, payloadlen); + goto out; + } + if (payloadlen != 0) { + err = nvlist_unpack(payload, payloadlen, &begin_nvl, KM_SLEEP); + kmem_free(payload, payloadlen); + if (err != 0) + goto out; + } + + if (featureflags & DMU_BACKUP_FEATURE_RESUMING) { + err = resume_check(&ra, begin_nvl); + if (err != 0) + goto out; + } + + (void) bqueue_init(&rwa.q, zfs_recv_queue_length, + offsetof(struct receive_record_arg, node)); + cv_init(&rwa.cv, NULL, CV_DEFAULT, NULL); + mutex_init(&rwa.mutex, NULL, MUTEX_DEFAULT, NULL); + rwa.os = ra.os; + rwa.byteswap = drc->drc_byteswap; + rwa.resumable = drc->drc_resumable; + + (void) thread_create(NULL, 0, receive_writer_thread, &rwa, 0, curproc, + TS_RUN, minclsyspri); + /* + * We're reading rwa.err without locks, which is safe since we are the + * only reader, and the worker thread is the only writer. It's ok if we + * miss a write for an iteration or two of the loop, since the writer + * thread will keep freeing records we send it until we send it an eos + * marker. + * + * We can leave this loop in 3 ways: First, if rwa.err is + * non-zero. In that case, the writer thread will free the rrd we just + * pushed. Second, if we're interrupted; in that case, either it's the + * first loop and ra.rrd was never allocated, or it's later, and ra.rrd + * has been handed off to the writer thread who will free it. Finally, + * if receive_read_record fails or we're at the end of the stream, then + * we free ra.rrd and exit. + */ + while (rwa.err == 0) { + if (issig(JUSTLOOKING) && issig(FORREAL)) { + err = SET_ERROR(EINTR); + break; + } + + ASSERT3P(ra.rrd, ==, NULL); + ra.rrd = ra.next_rrd; + ra.next_rrd = NULL; + /* Allocates and loads header into ra.next_rrd */ + err = receive_read_record(&ra); + + if (ra.rrd->header.drr_type == DRR_END || err != 0) { + kmem_free(ra.rrd, sizeof (*ra.rrd)); + ra.rrd = NULL; + break; + } + + bqueue_enqueue(&rwa.q, ra.rrd, + sizeof (struct receive_record_arg) + ra.rrd->payload_size); + ra.rrd = NULL; + } + if (ra.next_rrd == NULL) + ra.next_rrd = kmem_zalloc(sizeof (*ra.next_rrd), KM_SLEEP); + ra.next_rrd->eos_marker = B_TRUE; + bqueue_enqueue(&rwa.q, ra.next_rrd, 1); + + mutex_enter(&rwa.mutex); + while (!rwa.done) { + cv_wait(&rwa.cv, &rwa.mutex); + } + mutex_exit(&rwa.mutex); + + /* + * If we are receiving a full stream as a clone, all object IDs which + * are greater than the maximum ID referenced in the stream are + * by definition unused and must be freed. Note that it's possible that + * we've resumed this send and the first record we received was the END + * record. In that case, max_object would be 0, but we shouldn't start + * freeing all objects from there; instead we should start from the + * resumeobj. + */ + if (drc->drc_clone && drc->drc_drrb->drr_fromguid == 0) { + uint64_t obj; + if (nvlist_lookup_uint64(begin_nvl, "resume_object", &obj) != 0) + obj = 0; + if (rwa.max_object > obj) + obj = rwa.max_object; + obj++; + int free_err = 0; + int next_err = 0; + + while (next_err == 0) { + free_err = dmu_free_long_object(rwa.os, obj); + if (free_err != 0 && free_err != ENOENT) + break; + + next_err = dmu_object_next(rwa.os, &obj, FALSE, 0); + } + + if (err == 0) { + if (free_err != 0 && free_err != ENOENT) + err = free_err; + else if (next_err != ESRCH) + err = next_err; + } + } + + cv_destroy(&rwa.cv); + mutex_destroy(&rwa.mutex); + bqueue_destroy(&rwa.q); + if (err == 0) + err = rwa.err; + +out: + nvlist_free(begin_nvl); + if ((featureflags & DMU_BACKUP_FEATURE_DEDUP) && (cleanup_fd != -1)) + zfs_onexit_fd_rele(cleanup_fd); + + if (err != 0) { + /* + * Clean up references. If receive is not resumable, + * destroy what we created, so we don't leave it in + * the inconsistent state. + */ + dmu_recv_cleanup_ds(drc); + } + + *voffp = ra.voff; + objlist_destroy(&ra.ignore_objlist); + return (err); +} + +static int +dmu_recv_end_check(void *arg, dmu_tx_t *tx) +{ + dmu_recv_cookie_t *drc = arg; + dsl_pool_t *dp = dmu_tx_pool(tx); + int error; + + ASSERT3P(drc->drc_ds->ds_owner, ==, dmu_recv_tag); + + if (!drc->drc_newfs) { + dsl_dataset_t *origin_head; + + error = dsl_dataset_hold(dp, drc->drc_tofs, FTAG, &origin_head); + if (error != 0) + return (error); + if (drc->drc_force) { + /* + * We will destroy any snapshots in tofs (i.e. before + * origin_head) that are after the origin (which is + * the snap before drc_ds, because drc_ds can not + * have any snaps of its own). + */ + uint64_t obj; + + obj = dsl_dataset_phys(origin_head)->ds_prev_snap_obj; + while (obj != + dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj) { + dsl_dataset_t *snap; + error = dsl_dataset_hold_obj(dp, obj, FTAG, + &snap); + if (error != 0) + break; + if (snap->ds_dir != origin_head->ds_dir) + error = SET_ERROR(EINVAL); + if (error == 0) { + error = dsl_destroy_snapshot_check_impl( + snap, B_FALSE); + } + obj = dsl_dataset_phys(snap)->ds_prev_snap_obj; + dsl_dataset_rele(snap, FTAG); + if (error != 0) + break; + } + if (error != 0) { + dsl_dataset_rele(origin_head, FTAG); + return (error); + } + } + error = dsl_dataset_clone_swap_check_impl(drc->drc_ds, + origin_head, drc->drc_force, drc->drc_owner, tx); + if (error != 0) { + dsl_dataset_rele(origin_head, FTAG); + return (error); + } + error = dsl_dataset_snapshot_check_impl(origin_head, + drc->drc_tosnap, tx, B_TRUE, 1, drc->drc_cred); + dsl_dataset_rele(origin_head, FTAG); + if (error != 0) + return (error); + + error = dsl_destroy_head_check_impl(drc->drc_ds, 1); + } else { + error = dsl_dataset_snapshot_check_impl(drc->drc_ds, + drc->drc_tosnap, tx, B_TRUE, 1, drc->drc_cred); + } + return (error); +} + +static void +dmu_recv_end_sync(void *arg, dmu_tx_t *tx) +{ + dmu_recv_cookie_t *drc = arg; + dsl_pool_t *dp = dmu_tx_pool(tx); + + spa_history_log_internal_ds(drc->drc_ds, "finish receiving", + tx, "snap=%s", drc->drc_tosnap); + + if (!drc->drc_newfs) { + dsl_dataset_t *origin_head; + + VERIFY0(dsl_dataset_hold(dp, drc->drc_tofs, FTAG, + &origin_head)); + + if (drc->drc_force) { + /* + * Destroy any snapshots of drc_tofs (origin_head) + * after the origin (the snap before drc_ds). + */ + uint64_t obj; + + obj = dsl_dataset_phys(origin_head)->ds_prev_snap_obj; + while (obj != + dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj) { + dsl_dataset_t *snap; + VERIFY0(dsl_dataset_hold_obj(dp, obj, FTAG, + &snap)); + ASSERT3P(snap->ds_dir, ==, origin_head->ds_dir); + obj = dsl_dataset_phys(snap)->ds_prev_snap_obj; + dsl_destroy_snapshot_sync_impl(snap, + B_FALSE, tx); + dsl_dataset_rele(snap, FTAG); + } + } + VERIFY3P(drc->drc_ds->ds_prev, ==, + origin_head->ds_prev); + + dsl_dataset_clone_swap_sync_impl(drc->drc_ds, + origin_head, tx); + dsl_dataset_snapshot_sync_impl(origin_head, + drc->drc_tosnap, tx); + + /* set snapshot's creation time and guid */ + dmu_buf_will_dirty(origin_head->ds_prev->ds_dbuf, tx); + dsl_dataset_phys(origin_head->ds_prev)->ds_creation_time = + drc->drc_drrb->drr_creation_time; + dsl_dataset_phys(origin_head->ds_prev)->ds_guid = + drc->drc_drrb->drr_toguid; + dsl_dataset_phys(origin_head->ds_prev)->ds_flags &= + ~DS_FLAG_INCONSISTENT; + + dmu_buf_will_dirty(origin_head->ds_dbuf, tx); + dsl_dataset_phys(origin_head)->ds_flags &= + ~DS_FLAG_INCONSISTENT; + + drc->drc_newsnapobj = + dsl_dataset_phys(origin_head)->ds_prev_snap_obj; + + dsl_dataset_rele(origin_head, FTAG); + dsl_destroy_head_sync_impl(drc->drc_ds, tx); + + if (drc->drc_owner != NULL) + VERIFY3P(origin_head->ds_owner, ==, drc->drc_owner); + } else { + dsl_dataset_t *ds = drc->drc_ds; + + dsl_dataset_snapshot_sync_impl(ds, drc->drc_tosnap, tx); + + /* set snapshot's creation time and guid */ + dmu_buf_will_dirty(ds->ds_prev->ds_dbuf, tx); + dsl_dataset_phys(ds->ds_prev)->ds_creation_time = + drc->drc_drrb->drr_creation_time; + dsl_dataset_phys(ds->ds_prev)->ds_guid = + drc->drc_drrb->drr_toguid; + dsl_dataset_phys(ds->ds_prev)->ds_flags &= + ~DS_FLAG_INCONSISTENT; + + dmu_buf_will_dirty(ds->ds_dbuf, tx); + dsl_dataset_phys(ds)->ds_flags &= ~DS_FLAG_INCONSISTENT; + if (dsl_dataset_has_resume_receive_state(ds)) { + (void) zap_remove(dp->dp_meta_objset, ds->ds_object, + DS_FIELD_RESUME_FROMGUID, tx); + (void) zap_remove(dp->dp_meta_objset, ds->ds_object, + DS_FIELD_RESUME_OBJECT, tx); + (void) zap_remove(dp->dp_meta_objset, ds->ds_object, + DS_FIELD_RESUME_OFFSET, tx); + (void) zap_remove(dp->dp_meta_objset, ds->ds_object, + DS_FIELD_RESUME_BYTES, tx); + (void) zap_remove(dp->dp_meta_objset, ds->ds_object, + DS_FIELD_RESUME_TOGUID, tx); + (void) zap_remove(dp->dp_meta_objset, ds->ds_object, + DS_FIELD_RESUME_TONAME, tx); + } + drc->drc_newsnapobj = + dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj; + } + /* + * Release the hold from dmu_recv_begin. This must be done before + * we return to open context, so that when we free the dataset's dnode, + * we can evict its bonus buffer. + */ + dsl_dataset_disown(drc->drc_ds, dmu_recv_tag); + drc->drc_ds = NULL; +} + +static int +add_ds_to_guidmap(const char *name, avl_tree_t *guid_map, uint64_t snapobj) +{ + dsl_pool_t *dp; + dsl_dataset_t *snapds; + guid_map_entry_t *gmep; + int err; + + ASSERT(guid_map != NULL); + + err = dsl_pool_hold(name, FTAG, &dp); + if (err != 0) + return (err); + gmep = kmem_alloc(sizeof (*gmep), KM_SLEEP); + err = dsl_dataset_hold_obj(dp, snapobj, gmep, &snapds); + if (err == 0) { + gmep->guid = dsl_dataset_phys(snapds)->ds_guid; + gmep->gme_ds = snapds; + avl_add(guid_map, gmep); + dsl_dataset_long_hold(snapds, gmep); + } else { + kmem_free(gmep, sizeof (*gmep)); + } + + dsl_pool_rele(dp, FTAG); + return (err); +} + +static int dmu_recv_end_modified_blocks = 3; + +static int +dmu_recv_existing_end(dmu_recv_cookie_t *drc) +{ +#ifdef _KERNEL + /* + * We will be destroying the ds; make sure its origin is unmounted if + * necessary. + */ + char name[ZFS_MAX_DATASET_NAME_LEN]; + dsl_dataset_name(drc->drc_ds, name); + zfs_destroy_unmount_origin(name); +#endif + + return (dsl_sync_task(drc->drc_tofs, + dmu_recv_end_check, dmu_recv_end_sync, drc, + dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL)); +} + +static int +dmu_recv_new_end(dmu_recv_cookie_t *drc) +{ + return (dsl_sync_task(drc->drc_tofs, + dmu_recv_end_check, dmu_recv_end_sync, drc, + dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL)); +} + +int +dmu_recv_end(dmu_recv_cookie_t *drc, void *owner) +{ + int error; + + drc->drc_owner = owner; + + if (drc->drc_newfs) + error = dmu_recv_new_end(drc); + else + error = dmu_recv_existing_end(drc); + + if (error != 0) { + dmu_recv_cleanup_ds(drc); + } else if (drc->drc_guid_to_ds_map != NULL) { + (void) add_ds_to_guidmap(drc->drc_tofs, + drc->drc_guid_to_ds_map, + drc->drc_newsnapobj); + } + return (error); +} + +/* + * Return TRUE if this objset is currently being received into. + */ +boolean_t +dmu_objset_is_receiving(objset_t *os) +{ + return (os->os_dsl_dataset != NULL && + os->os_dsl_dataset->ds_owner == dmu_recv_tag); +} diff --git a/usr/src/uts/common/fs/zfs/dmu_send.c b/usr/src/uts/common/fs/zfs/dmu_send.c index f4b181f62a..d42a7c66de 100644 --- a/usr/src/uts/common/fs/zfs/dmu_send.c +++ b/usr/src/uts/common/fs/zfs/dmu_send.c @@ -59,13 +59,9 @@ /* Set this tunable to TRUE to replace corrupt data with 0x2f5baddb10c */ int zfs_send_corrupt_data = B_FALSE; int zfs_send_queue_length = 16 * 1024 * 1024; -int zfs_recv_queue_length = 16 * 1024 * 1024; /* Set this tunable to FALSE to disable setting of DRR_FLAG_FREERECORDS */ int zfs_send_set_freerecords_bit = B_TRUE; -static char *dmu_recv_tag = "dmu_recv_tag"; -const char *recv_clone_name = "%recv"; - /* * Use this to override the recordsize calculation for fast zfs send estimates. */ @@ -75,8 +71,6 @@ uint64_t zfs_override_estimate_recordsize = 0; (((uint64_t)datablkszsec) << (SPA_MINBLOCKSHIFT + \ (level) * (indblkshift - SPA_BLKPTRSHIFT))) -static void byteswap_record(dmu_replay_record_t *drr); - struct send_thread_arg { bqueue_t q; dsl_dataset_t *ds; /* Dataset to traverse */ @@ -1257,2203 +1251,3 @@ dmu_send_estimate_from_txg(dsl_dataset_t *ds, uint64_t from_txg, size.compressed, stream_compressed, sizep); return (err); } - -typedef struct dmu_recv_begin_arg { - const char *drba_origin; - dmu_recv_cookie_t *drba_cookie; - cred_t *drba_cred; - uint64_t drba_snapobj; -} dmu_recv_begin_arg_t; - -static int -recv_begin_check_existing_impl(dmu_recv_begin_arg_t *drba, dsl_dataset_t *ds, - uint64_t fromguid) -{ - uint64_t val; - int error; - dsl_pool_t *dp = ds->ds_dir->dd_pool; - - /* temporary clone name must not exist */ - error = zap_lookup(dp->dp_meta_objset, - dsl_dir_phys(ds->ds_dir)->dd_child_dir_zapobj, recv_clone_name, - 8, 1, &val); - if (error != ENOENT) - return (error == 0 ? EBUSY : error); - - /* new snapshot name must not exist */ - error = zap_lookup(dp->dp_meta_objset, - dsl_dataset_phys(ds)->ds_snapnames_zapobj, - drba->drba_cookie->drc_tosnap, 8, 1, &val); - if (error != ENOENT) - return (error == 0 ? EEXIST : error); - - /* - * Check snapshot limit before receiving. We'll recheck again at the - * end, but might as well abort before receiving if we're already over - * the limit. - * - * Note that we do not check the file system limit with - * dsl_dir_fscount_check because the temporary %clones don't count - * against that limit. - */ - error = dsl_fs_ss_limit_check(ds->ds_dir, 1, ZFS_PROP_SNAPSHOT_LIMIT, - NULL, drba->drba_cred); - if (error != 0) - return (error); - - if (fromguid != 0) { - dsl_dataset_t *snap; - uint64_t obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; - - /* Find snapshot in this dir that matches fromguid. */ - while (obj != 0) { - error = dsl_dataset_hold_obj(dp, obj, FTAG, - &snap); - if (error != 0) - return (SET_ERROR(ENODEV)); - if (snap->ds_dir != ds->ds_dir) { - dsl_dataset_rele(snap, FTAG); - return (SET_ERROR(ENODEV)); - } - if (dsl_dataset_phys(snap)->ds_guid == fromguid) - break; - obj = dsl_dataset_phys(snap)->ds_prev_snap_obj; - dsl_dataset_rele(snap, FTAG); - } - if (obj == 0) - return (SET_ERROR(ENODEV)); - - if (drba->drba_cookie->drc_force) { - drba->drba_snapobj = obj; - } else { - /* - * If we are not forcing, there must be no - * changes since fromsnap. - */ - if (dsl_dataset_modified_since_snap(ds, snap)) { - dsl_dataset_rele(snap, FTAG); - return (SET_ERROR(ETXTBSY)); - } - drba->drba_snapobj = ds->ds_prev->ds_object; - } - - dsl_dataset_rele(snap, FTAG); - } else { - /* if full, then must be forced */ - if (!drba->drba_cookie->drc_force) - return (SET_ERROR(EEXIST)); - /* start from $ORIGIN@$ORIGIN, if supported */ - drba->drba_snapobj = dp->dp_origin_snap != NULL ? - dp->dp_origin_snap->ds_object : 0; - } - - return (0); - -} - -static int -dmu_recv_begin_check(void *arg, dmu_tx_t *tx) -{ - dmu_recv_begin_arg_t *drba = arg; - dsl_pool_t *dp = dmu_tx_pool(tx); - struct drr_begin *drrb = drba->drba_cookie->drc_drrb; - uint64_t fromguid = drrb->drr_fromguid; - int flags = drrb->drr_flags; - int error; - uint64_t featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo); - dsl_dataset_t *ds; - const char *tofs = drba->drba_cookie->drc_tofs; - - /* already checked */ - ASSERT3U(drrb->drr_magic, ==, DMU_BACKUP_MAGIC); - ASSERT(!(featureflags & DMU_BACKUP_FEATURE_RESUMING)); - - if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == - DMU_COMPOUNDSTREAM || - drrb->drr_type >= DMU_OST_NUMTYPES || - ((flags & DRR_FLAG_CLONE) && drba->drba_origin == NULL)) - return (SET_ERROR(EINVAL)); - - /* Verify pool version supports SA if SA_SPILL feature set */ - if ((featureflags & DMU_BACKUP_FEATURE_SA_SPILL) && - spa_version(dp->dp_spa) < SPA_VERSION_SA) - return (SET_ERROR(ENOTSUP)); - - if (drba->drba_cookie->drc_resumable && - !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EXTENSIBLE_DATASET)) - return (SET_ERROR(ENOTSUP)); - - /* - * The receiving code doesn't know how to translate a WRITE_EMBEDDED - * record to a plain WRITE record, so the pool must have the - * EMBEDDED_DATA feature enabled if the stream has WRITE_EMBEDDED - * records. Same with WRITE_EMBEDDED records that use LZ4 compression. - */ - if ((featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) && - !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EMBEDDED_DATA)) - return (SET_ERROR(ENOTSUP)); - if ((featureflags & DMU_BACKUP_FEATURE_LZ4) && - !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LZ4_COMPRESS)) - return (SET_ERROR(ENOTSUP)); - - /* - * The receiving code doesn't know how to translate large blocks - * to smaller ones, so the pool must have the LARGE_BLOCKS - * feature enabled if the stream has LARGE_BLOCKS. Same with - * large dnodes. - */ - if ((featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) && - !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_BLOCKS)) - return (SET_ERROR(ENOTSUP)); - if ((featureflags & DMU_BACKUP_FEATURE_LARGE_DNODE) && - !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_DNODE)) - return (SET_ERROR(ENOTSUP)); - - error = dsl_dataset_hold(dp, tofs, FTAG, &ds); - if (error == 0) { - /* target fs already exists; recv into temp clone */ - - /* Can't recv a clone into an existing fs */ - if (flags & DRR_FLAG_CLONE || drba->drba_origin) { - dsl_dataset_rele(ds, FTAG); - return (SET_ERROR(EINVAL)); - } - - error = recv_begin_check_existing_impl(drba, ds, fromguid); - dsl_dataset_rele(ds, FTAG); - } else if (error == ENOENT) { - /* target fs does not exist; must be a full backup or clone */ - char buf[ZFS_MAX_DATASET_NAME_LEN]; - - /* - * If it's a non-clone incremental, we are missing the - * target fs, so fail the recv. - */ - if (fromguid != 0 && !(flags & DRR_FLAG_CLONE || - drba->drba_origin)) - return (SET_ERROR(ENOENT)); - - /* - * If we're receiving a full send as a clone, and it doesn't - * contain all the necessary free records and freeobject - * records, reject it. - */ - if (fromguid == 0 && drba->drba_origin && - !(flags & DRR_FLAG_FREERECORDS)) - return (SET_ERROR(EINVAL)); - - /* Open the parent of tofs */ - ASSERT3U(strlen(tofs), <, sizeof (buf)); - (void) strlcpy(buf, tofs, strrchr(tofs, '/') - tofs + 1); - error = dsl_dataset_hold(dp, buf, FTAG, &ds); - if (error != 0) - return (error); - - /* - * Check filesystem and snapshot limits before receiving. We'll - * recheck snapshot limits again at the end (we create the - * filesystems and increment those counts during begin_sync). - */ - error = dsl_fs_ss_limit_check(ds->ds_dir, 1, - ZFS_PROP_FILESYSTEM_LIMIT, NULL, drba->drba_cred); - if (error != 0) { - dsl_dataset_rele(ds, FTAG); - return (error); - } - - error = dsl_fs_ss_limit_check(ds->ds_dir, 1, - ZFS_PROP_SNAPSHOT_LIMIT, NULL, drba->drba_cred); - if (error != 0) { - dsl_dataset_rele(ds, FTAG); - return (error); - } - - if (drba->drba_origin != NULL) { - dsl_dataset_t *origin; - error = dsl_dataset_hold(dp, drba->drba_origin, - FTAG, &origin); - if (error != 0) { - dsl_dataset_rele(ds, FTAG); - return (error); - } - if (!origin->ds_is_snapshot) { - dsl_dataset_rele(origin, FTAG); - dsl_dataset_rele(ds, FTAG); - return (SET_ERROR(EINVAL)); - } - if (dsl_dataset_phys(origin)->ds_guid != fromguid && - fromguid != 0) { - dsl_dataset_rele(origin, FTAG); - dsl_dataset_rele(ds, FTAG); - return (SET_ERROR(ENODEV)); - } - dsl_dataset_rele(origin, FTAG); - } - dsl_dataset_rele(ds, FTAG); - error = 0; - } - return (error); -} - -static void -dmu_recv_begin_sync(void *arg, dmu_tx_t *tx) -{ - dmu_recv_begin_arg_t *drba = arg; - dsl_pool_t *dp = dmu_tx_pool(tx); - objset_t *mos = dp->dp_meta_objset; - struct drr_begin *drrb = drba->drba_cookie->drc_drrb; - const char *tofs = drba->drba_cookie->drc_tofs; - dsl_dataset_t *ds, *newds; - uint64_t dsobj; - int error; - uint64_t crflags = 0; - - if (drrb->drr_flags & DRR_FLAG_CI_DATA) - crflags |= DS_FLAG_CI_DATASET; - - error = dsl_dataset_hold(dp, tofs, FTAG, &ds); - if (error == 0) { - /* create temporary clone */ - dsl_dataset_t *snap = NULL; - if (drba->drba_snapobj != 0) { - VERIFY0(dsl_dataset_hold_obj(dp, - drba->drba_snapobj, FTAG, &snap)); - } - dsobj = dsl_dataset_create_sync(ds->ds_dir, recv_clone_name, - snap, crflags, drba->drba_cred, tx); - if (drba->drba_snapobj != 0) - dsl_dataset_rele(snap, FTAG); - dsl_dataset_rele(ds, FTAG); - } else { - dsl_dir_t *dd; - const char *tail; - dsl_dataset_t *origin = NULL; - - VERIFY0(dsl_dir_hold(dp, tofs, FTAG, &dd, &tail)); - - if (drba->drba_origin != NULL) { - VERIFY0(dsl_dataset_hold(dp, drba->drba_origin, - FTAG, &origin)); - } - - /* Create new dataset. */ - dsobj = dsl_dataset_create_sync(dd, - strrchr(tofs, '/') + 1, - origin, crflags, drba->drba_cred, tx); - if (origin != NULL) - dsl_dataset_rele(origin, FTAG); - dsl_dir_rele(dd, FTAG); - drba->drba_cookie->drc_newfs = B_TRUE; - } - VERIFY0(dsl_dataset_own_obj(dp, dsobj, dmu_recv_tag, &newds)); - - if (drba->drba_cookie->drc_resumable) { - dsl_dataset_zapify(newds, tx); - if (drrb->drr_fromguid != 0) { - VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_FROMGUID, - 8, 1, &drrb->drr_fromguid, tx)); - } - VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_TOGUID, - 8, 1, &drrb->drr_toguid, tx)); - VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_TONAME, - 1, strlen(drrb->drr_toname) + 1, drrb->drr_toname, tx)); - uint64_t one = 1; - uint64_t zero = 0; - VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_OBJECT, - 8, 1, &one, tx)); - VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_OFFSET, - 8, 1, &zero, tx)); - VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_BYTES, - 8, 1, &zero, tx)); - if (DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo) & - DMU_BACKUP_FEATURE_LARGE_BLOCKS) { - VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_LARGEBLOCK, - 8, 1, &one, tx)); - } - if (DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo) & - DMU_BACKUP_FEATURE_EMBED_DATA) { - VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_EMBEDOK, - 8, 1, &one, tx)); - } - if (DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo) & - DMU_BACKUP_FEATURE_COMPRESSED) { - VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_COMPRESSOK, - 8, 1, &one, tx)); - } - } - - dmu_buf_will_dirty(newds->ds_dbuf, tx); - dsl_dataset_phys(newds)->ds_flags |= DS_FLAG_INCONSISTENT; - - /* - * If we actually created a non-clone, we need to create the - * objset in our new dataset. - */ - rrw_enter(&newds->ds_bp_rwlock, RW_READER, FTAG); - if (BP_IS_HOLE(dsl_dataset_get_blkptr(newds))) { - (void) dmu_objset_create_impl(dp->dp_spa, - newds, dsl_dataset_get_blkptr(newds), drrb->drr_type, tx); - } - rrw_exit(&newds->ds_bp_rwlock, FTAG); - - drba->drba_cookie->drc_ds = newds; - - spa_history_log_internal_ds(newds, "receive", tx, ""); -} - -static int -dmu_recv_resume_begin_check(void *arg, dmu_tx_t *tx) -{ - dmu_recv_begin_arg_t *drba = arg; - dsl_pool_t *dp = dmu_tx_pool(tx); - struct drr_begin *drrb = drba->drba_cookie->drc_drrb; - int error; - uint64_t featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo); - dsl_dataset_t *ds; - const char *tofs = drba->drba_cookie->drc_tofs; - - /* 6 extra bytes for /%recv */ - char recvname[ZFS_MAX_DATASET_NAME_LEN + 6]; - - /* already checked */ - ASSERT3U(drrb->drr_magic, ==, DMU_BACKUP_MAGIC); - ASSERT(featureflags & DMU_BACKUP_FEATURE_RESUMING); - - if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == - DMU_COMPOUNDSTREAM || - drrb->drr_type >= DMU_OST_NUMTYPES) - return (SET_ERROR(EINVAL)); - - /* Verify pool version supports SA if SA_SPILL feature set */ - if ((featureflags & DMU_BACKUP_FEATURE_SA_SPILL) && - spa_version(dp->dp_spa) < SPA_VERSION_SA) - return (SET_ERROR(ENOTSUP)); - - /* - * The receiving code doesn't know how to translate a WRITE_EMBEDDED - * record to a plain WRITE record, so the pool must have the - * EMBEDDED_DATA feature enabled if the stream has WRITE_EMBEDDED - * records. Same with WRITE_EMBEDDED records that use LZ4 compression. - */ - if ((featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) && - !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EMBEDDED_DATA)) - return (SET_ERROR(ENOTSUP)); - if ((featureflags & DMU_BACKUP_FEATURE_LZ4) && - !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LZ4_COMPRESS)) - return (SET_ERROR(ENOTSUP)); - - /* - * The receiving code doesn't know how to translate large blocks - * to smaller ones, so the pool must have the LARGE_BLOCKS - * feature enabled if the stream has LARGE_BLOCKS. Same with - * large dnodes. - */ - if ((featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) && - !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_BLOCKS)) - return (SET_ERROR(ENOTSUP)); - if ((featureflags & DMU_BACKUP_FEATURE_LARGE_DNODE) && - !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_DNODE)) - return (SET_ERROR(ENOTSUP)); - - (void) snprintf(recvname, sizeof (recvname), "%s/%s", - tofs, recv_clone_name); - - if (dsl_dataset_hold(dp, recvname, FTAG, &ds) != 0) { - /* %recv does not exist; continue in tofs */ - error = dsl_dataset_hold(dp, tofs, FTAG, &ds); - if (error != 0) - return (error); - } - - /* check that ds is marked inconsistent */ - if (!DS_IS_INCONSISTENT(ds)) { - dsl_dataset_rele(ds, FTAG); - return (SET_ERROR(EINVAL)); - } - - /* check that there is resuming data, and that the toguid matches */ - if (!dsl_dataset_is_zapified(ds)) { - dsl_dataset_rele(ds, FTAG); - return (SET_ERROR(EINVAL)); - } - uint64_t val; - error = zap_lookup(dp->dp_meta_objset, ds->ds_object, - DS_FIELD_RESUME_TOGUID, sizeof (val), 1, &val); - if (error != 0 || drrb->drr_toguid != val) { - dsl_dataset_rele(ds, FTAG); - return (SET_ERROR(EINVAL)); - } - - /* - * Check if the receive is still running. If so, it will be owned. - * Note that nothing else can own the dataset (e.g. after the receive - * fails) because it will be marked inconsistent. - */ - if (dsl_dataset_has_owner(ds)) { - dsl_dataset_rele(ds, FTAG); - return (SET_ERROR(EBUSY)); - } - - /* There should not be any snapshots of this fs yet. */ - if (ds->ds_prev != NULL && ds->ds_prev->ds_dir == ds->ds_dir) { - dsl_dataset_rele(ds, FTAG); - return (SET_ERROR(EINVAL)); - } - - /* - * Note: resume point will be checked when we process the first WRITE - * record. - */ - - /* check that the origin matches */ - val = 0; - (void) zap_lookup(dp->dp_meta_objset, ds->ds_object, - DS_FIELD_RESUME_FROMGUID, sizeof (val), 1, &val); - if (drrb->drr_fromguid != val) { - dsl_dataset_rele(ds, FTAG); - return (SET_ERROR(EINVAL)); - } - - dsl_dataset_rele(ds, FTAG); - return (0); -} - -static void -dmu_recv_resume_begin_sync(void *arg, dmu_tx_t *tx) -{ - dmu_recv_begin_arg_t *drba = arg; - dsl_pool_t *dp = dmu_tx_pool(tx); - const char *tofs = drba->drba_cookie->drc_tofs; - dsl_dataset_t *ds; - uint64_t dsobj; - /* 6 extra bytes for /%recv */ - char recvname[ZFS_MAX_DATASET_NAME_LEN + 6]; - - (void) snprintf(recvname, sizeof (recvname), "%s/%s", - tofs, recv_clone_name); - - if (dsl_dataset_hold(dp, recvname, FTAG, &ds) != 0) { - /* %recv does not exist; continue in tofs */ - VERIFY0(dsl_dataset_hold(dp, tofs, FTAG, &ds)); - drba->drba_cookie->drc_newfs = B_TRUE; - } - - /* clear the inconsistent flag so that we can own it */ - ASSERT(DS_IS_INCONSISTENT(ds)); - dmu_buf_will_dirty(ds->ds_dbuf, tx); - dsl_dataset_phys(ds)->ds_flags &= ~DS_FLAG_INCONSISTENT; - dsobj = ds->ds_object; - dsl_dataset_rele(ds, FTAG); - - VERIFY0(dsl_dataset_own_obj(dp, dsobj, dmu_recv_tag, &ds)); - - dmu_buf_will_dirty(ds->ds_dbuf, tx); - dsl_dataset_phys(ds)->ds_flags |= DS_FLAG_INCONSISTENT; - - rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG); - ASSERT(!BP_IS_HOLE(dsl_dataset_get_blkptr(ds))); - rrw_exit(&ds->ds_bp_rwlock, FTAG); - - drba->drba_cookie->drc_ds = ds; - - spa_history_log_internal_ds(ds, "resume receive", tx, ""); -} - -/* - * NB: callers *MUST* call dmu_recv_stream() if dmu_recv_begin() - * succeeds; otherwise we will leak the holds on the datasets. - */ -int -dmu_recv_begin(char *tofs, char *tosnap, dmu_replay_record_t *drr_begin, - boolean_t force, boolean_t resumable, char *origin, dmu_recv_cookie_t *drc) -{ - dmu_recv_begin_arg_t drba = { 0 }; - - bzero(drc, sizeof (dmu_recv_cookie_t)); - drc->drc_drr_begin = drr_begin; - drc->drc_drrb = &drr_begin->drr_u.drr_begin; - drc->drc_tosnap = tosnap; - drc->drc_tofs = tofs; - drc->drc_force = force; - drc->drc_resumable = resumable; - drc->drc_cred = CRED(); - drc->drc_clone = (origin != NULL); - - if (drc->drc_drrb->drr_magic == BSWAP_64(DMU_BACKUP_MAGIC)) { - drc->drc_byteswap = B_TRUE; - (void) fletcher_4_incremental_byteswap(drr_begin, - sizeof (dmu_replay_record_t), &drc->drc_cksum); - byteswap_record(drr_begin); - } else if (drc->drc_drrb->drr_magic == DMU_BACKUP_MAGIC) { - (void) fletcher_4_incremental_native(drr_begin, - sizeof (dmu_replay_record_t), &drc->drc_cksum); - } else { - return (SET_ERROR(EINVAL)); - } - - drba.drba_origin = origin; - drba.drba_cookie = drc; - drba.drba_cred = CRED(); - - if (DMU_GET_FEATUREFLAGS(drc->drc_drrb->drr_versioninfo) & - DMU_BACKUP_FEATURE_RESUMING) { - return (dsl_sync_task(tofs, - dmu_recv_resume_begin_check, dmu_recv_resume_begin_sync, - &drba, 5, ZFS_SPACE_CHECK_NORMAL)); - } else { - return (dsl_sync_task(tofs, - dmu_recv_begin_check, dmu_recv_begin_sync, - &drba, 5, ZFS_SPACE_CHECK_NORMAL)); - } -} - -struct receive_record_arg { - dmu_replay_record_t header; - void *payload; /* Pointer to a buffer containing the payload */ - /* - * If the record is a write, pointer to the arc_buf_t containing the - * payload. - */ - arc_buf_t *write_buf; - int payload_size; - uint64_t bytes_read; /* bytes read from stream when record created */ - boolean_t eos_marker; /* Marks the end of the stream */ - bqueue_node_t node; -}; - -struct receive_writer_arg { - objset_t *os; - boolean_t byteswap; - bqueue_t q; - - /* - * These three args are used to signal to the main thread that we're - * done. - */ - kmutex_t mutex; - kcondvar_t cv; - boolean_t done; - - int err; - /* A map from guid to dataset to help handle dedup'd streams. */ - avl_tree_t *guid_to_ds_map; - boolean_t resumable; - uint64_t last_object; - uint64_t last_offset; - uint64_t max_object; /* highest object ID referenced in stream */ - uint64_t bytes_read; /* bytes read when current record created */ -}; - -struct objlist { - list_t list; /* List of struct receive_objnode. */ - /* - * Last object looked up. Used to assert that objects are being looked - * up in ascending order. - */ - uint64_t last_lookup; -}; - -struct receive_objnode { - list_node_t node; - uint64_t object; -}; - -struct receive_arg { - objset_t *os; - vnode_t *vp; /* The vnode to read the stream from */ - uint64_t voff; /* The current offset in the stream */ - uint64_t bytes_read; - /* - * A record that has had its payload read in, but hasn't yet been handed - * off to the worker thread. - */ - struct receive_record_arg *rrd; - /* A record that has had its header read in, but not its payload. */ - struct receive_record_arg *next_rrd; - zio_cksum_t cksum; - zio_cksum_t prev_cksum; - int err; - boolean_t byteswap; - /* Sorted list of objects not to issue prefetches for. */ - struct objlist ignore_objlist; -}; - -typedef struct guid_map_entry { - uint64_t guid; - dsl_dataset_t *gme_ds; - avl_node_t avlnode; -} guid_map_entry_t; - -static int -guid_compare(const void *arg1, const void *arg2) -{ - const guid_map_entry_t *gmep1 = arg1; - const guid_map_entry_t *gmep2 = arg2; - - if (gmep1->guid < gmep2->guid) - return (-1); - else if (gmep1->guid > gmep2->guid) - return (1); - return (0); -} - -static void -free_guid_map_onexit(void *arg) -{ - avl_tree_t *ca = arg; - void *cookie = NULL; - guid_map_entry_t *gmep; - - while ((gmep = avl_destroy_nodes(ca, &cookie)) != NULL) { - dsl_dataset_long_rele(gmep->gme_ds, gmep); - dsl_dataset_rele(gmep->gme_ds, gmep); - kmem_free(gmep, sizeof (guid_map_entry_t)); - } - avl_destroy(ca); - kmem_free(ca, sizeof (avl_tree_t)); -} - -static int -receive_read(struct receive_arg *ra, int len, void *buf) -{ - int done = 0; - - /* - * The code doesn't rely on this (lengths being multiples of 8). See - * comment in dump_bytes. - */ - ASSERT0(len % 8); - - while (done < len) { - ssize_t resid; - - ra->err = vn_rdwr(UIO_READ, ra->vp, - (char *)buf + done, len - done, - ra->voff, UIO_SYSSPACE, FAPPEND, - RLIM64_INFINITY, CRED(), &resid); - - if (resid == len - done) { - /* - * Note: ECKSUM indicates that the receive - * was interrupted and can potentially be resumed. - */ - ra->err = SET_ERROR(ECKSUM); - } - ra->voff += len - done - resid; - done = len - resid; - if (ra->err != 0) - return (ra->err); - } - - ra->bytes_read += len; - - ASSERT3U(done, ==, len); - return (0); -} - -static void -byteswap_record(dmu_replay_record_t *drr) -{ -#define DO64(X) (drr->drr_u.X = BSWAP_64(drr->drr_u.X)) -#define DO32(X) (drr->drr_u.X = BSWAP_32(drr->drr_u.X)) - drr->drr_type = BSWAP_32(drr->drr_type); - drr->drr_payloadlen = BSWAP_32(drr->drr_payloadlen); - - switch (drr->drr_type) { - case DRR_BEGIN: - DO64(drr_begin.drr_magic); - DO64(drr_begin.drr_versioninfo); - DO64(drr_begin.drr_creation_time); - DO32(drr_begin.drr_type); - DO32(drr_begin.drr_flags); - DO64(drr_begin.drr_toguid); - DO64(drr_begin.drr_fromguid); - break; - case DRR_OBJECT: - DO64(drr_object.drr_object); - DO32(drr_object.drr_type); - DO32(drr_object.drr_bonustype); - DO32(drr_object.drr_blksz); - DO32(drr_object.drr_bonuslen); - DO64(drr_object.drr_toguid); - break; - case DRR_FREEOBJECTS: - DO64(drr_freeobjects.drr_firstobj); - DO64(drr_freeobjects.drr_numobjs); - DO64(drr_freeobjects.drr_toguid); - break; - case DRR_WRITE: - DO64(drr_write.drr_object); - DO32(drr_write.drr_type); - DO64(drr_write.drr_offset); - DO64(drr_write.drr_logical_size); - DO64(drr_write.drr_toguid); - ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_write.drr_key.ddk_cksum); - DO64(drr_write.drr_key.ddk_prop); - DO64(drr_write.drr_compressed_size); - break; - case DRR_WRITE_BYREF: - DO64(drr_write_byref.drr_object); - DO64(drr_write_byref.drr_offset); - DO64(drr_write_byref.drr_length); - DO64(drr_write_byref.drr_toguid); - DO64(drr_write_byref.drr_refguid); - DO64(drr_write_byref.drr_refobject); - DO64(drr_write_byref.drr_refoffset); - ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_write_byref. - drr_key.ddk_cksum); - DO64(drr_write_byref.drr_key.ddk_prop); - break; - case DRR_WRITE_EMBEDDED: - DO64(drr_write_embedded.drr_object); - DO64(drr_write_embedded.drr_offset); - DO64(drr_write_embedded.drr_length); - DO64(drr_write_embedded.drr_toguid); - DO32(drr_write_embedded.drr_lsize); - DO32(drr_write_embedded.drr_psize); - break; - case DRR_FREE: - DO64(drr_free.drr_object); - DO64(drr_free.drr_offset); - DO64(drr_free.drr_length); - DO64(drr_free.drr_toguid); - break; - case DRR_SPILL: - DO64(drr_spill.drr_object); - DO64(drr_spill.drr_length); - DO64(drr_spill.drr_toguid); - break; - case DRR_END: - DO64(drr_end.drr_toguid); - ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_end.drr_checksum); - break; - } - - if (drr->drr_type != DRR_BEGIN) { - ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_checksum.drr_checksum); - } - -#undef DO64 -#undef DO32 -} - -static inline uint8_t -deduce_nblkptr(dmu_object_type_t bonus_type, uint64_t bonus_size) -{ - if (bonus_type == DMU_OT_SA) { - return (1); - } else { - return (1 + - ((DN_OLD_MAX_BONUSLEN - - MIN(DN_OLD_MAX_BONUSLEN, bonus_size)) >> SPA_BLKPTRSHIFT)); - } -} - -static void -save_resume_state(struct receive_writer_arg *rwa, - uint64_t object, uint64_t offset, dmu_tx_t *tx) -{ - int txgoff = dmu_tx_get_txg(tx) & TXG_MASK; - - if (!rwa->resumable) - return; - - /* - * We use ds_resume_bytes[] != 0 to indicate that we need to - * update this on disk, so it must not be 0. - */ - ASSERT(rwa->bytes_read != 0); - - /* - * We only resume from write records, which have a valid - * (non-meta-dnode) object number. - */ - ASSERT(object != 0); - - /* - * For resuming to work correctly, we must receive records in order, - * sorted by object,offset. This is checked by the callers, but - * assert it here for good measure. - */ - ASSERT3U(object, >=, rwa->os->os_dsl_dataset->ds_resume_object[txgoff]); - ASSERT(object != rwa->os->os_dsl_dataset->ds_resume_object[txgoff] || - offset >= rwa->os->os_dsl_dataset->ds_resume_offset[txgoff]); - ASSERT3U(rwa->bytes_read, >=, - rwa->os->os_dsl_dataset->ds_resume_bytes[txgoff]); - - rwa->os->os_dsl_dataset->ds_resume_object[txgoff] = object; - rwa->os->os_dsl_dataset->ds_resume_offset[txgoff] = offset; - rwa->os->os_dsl_dataset->ds_resume_bytes[txgoff] = rwa->bytes_read; -} - -static int -receive_object(struct receive_writer_arg *rwa, struct drr_object *drro, - void *data) -{ - dmu_object_info_t doi; - dmu_tx_t *tx; - uint64_t object; - int err; - uint8_t dn_slots = drro->drr_dn_slots != 0 ? - drro->drr_dn_slots : DNODE_MIN_SLOTS; - - if (drro->drr_type == DMU_OT_NONE || - !DMU_OT_IS_VALID(drro->drr_type) || - !DMU_OT_IS_VALID(drro->drr_bonustype) || - drro->drr_checksumtype >= ZIO_CHECKSUM_FUNCTIONS || - drro->drr_compress >= ZIO_COMPRESS_FUNCTIONS || - P2PHASE(drro->drr_blksz, SPA_MINBLOCKSIZE) || - drro->drr_blksz < SPA_MINBLOCKSIZE || - drro->drr_blksz > spa_maxblocksize(dmu_objset_spa(rwa->os)) || - drro->drr_bonuslen > - DN_BONUS_SIZE(spa_maxdnodesize(dmu_objset_spa(rwa->os))) || - dn_slots > - (spa_maxdnodesize(dmu_objset_spa(rwa->os)) >> DNODE_SHIFT)) { - return (SET_ERROR(EINVAL)); - } - - err = dmu_object_info(rwa->os, drro->drr_object, &doi); - - if (err != 0 && err != ENOENT && err != EEXIST) - return (SET_ERROR(EINVAL)); - - if (drro->drr_object > rwa->max_object) - rwa->max_object = drro->drr_object; - - /* - * If we are losing blkptrs or changing the block size this must - * be a new file instance. We must clear out the previous file - * contents before we can change this type of metadata in the dnode. - */ - if (err == 0) { - int nblkptr; - - object = drro->drr_object; - - nblkptr = deduce_nblkptr(drro->drr_bonustype, - drro->drr_bonuslen); - - if (drro->drr_blksz != doi.doi_data_block_size || - nblkptr < doi.doi_nblkptr || - dn_slots != doi.doi_dnodesize >> DNODE_SHIFT) { - err = dmu_free_long_range(rwa->os, drro->drr_object, - 0, DMU_OBJECT_END); - if (err != 0) - return (SET_ERROR(EINVAL)); - } - } else if (err == EEXIST) { - /* - * The object requested is currently an interior slot of a - * multi-slot dnode. This will be resolved when the next txg - * is synced out, since the send stream will have told us - * to free this slot when we freed the associated dnode - * earlier in the stream. - */ - txg_wait_synced(dmu_objset_pool(rwa->os), 0); - object = drro->drr_object; - } else { - /* object is free and we are about to allocate a new one */ - object = DMU_NEW_OBJECT; - } - - /* - * If this is a multi-slot dnode there is a chance that this - * object will expand into a slot that is already used by - * another object from the previous snapshot. We must free - * these objects before we attempt to allocate the new dnode. - */ - if (dn_slots > 1) { - boolean_t need_sync = B_FALSE; - - for (uint64_t slot = drro->drr_object + 1; - slot < drro->drr_object + dn_slots; - slot++) { - dmu_object_info_t slot_doi; - - err = dmu_object_info(rwa->os, slot, &slot_doi); - if (err == ENOENT || err == EEXIST) - continue; - else if (err != 0) - return (err); - - err = dmu_free_long_object(rwa->os, slot); - - if (err != 0) - return (err); - - need_sync = B_TRUE; - } - - if (need_sync) - txg_wait_synced(dmu_objset_pool(rwa->os), 0); - } - - tx = dmu_tx_create(rwa->os); - dmu_tx_hold_bonus(tx, object); - err = dmu_tx_assign(tx, TXG_WAIT); - if (err != 0) { - dmu_tx_abort(tx); - return (err); - } - - if (object == DMU_NEW_OBJECT) { - /* currently free, want to be allocated */ - err = dmu_object_claim_dnsize(rwa->os, drro->drr_object, - drro->drr_type, drro->drr_blksz, - drro->drr_bonustype, drro->drr_bonuslen, - dn_slots << DNODE_SHIFT, tx); - } else if (drro->drr_type != doi.doi_type || - drro->drr_blksz != doi.doi_data_block_size || - drro->drr_bonustype != doi.doi_bonus_type || - drro->drr_bonuslen != doi.doi_bonus_size || - drro->drr_dn_slots != (doi.doi_dnodesize >> DNODE_SHIFT)) { - /* currently allocated, but with different properties */ - err = dmu_object_reclaim_dnsize(rwa->os, drro->drr_object, - drro->drr_type, drro->drr_blksz, - drro->drr_bonustype, drro->drr_bonuslen, - drro->drr_dn_slots << DNODE_SHIFT, tx); - } - if (err != 0) { - dmu_tx_commit(tx); - return (SET_ERROR(EINVAL)); - } - - dmu_object_set_checksum(rwa->os, drro->drr_object, - drro->drr_checksumtype, tx); - dmu_object_set_compress(rwa->os, drro->drr_object, - drro->drr_compress, tx); - - if (data != NULL) { - dmu_buf_t *db; - - VERIFY0(dmu_bonus_hold(rwa->os, drro->drr_object, FTAG, &db)); - dmu_buf_will_dirty(db, tx); - - ASSERT3U(db->db_size, >=, drro->drr_bonuslen); - bcopy(data, db->db_data, drro->drr_bonuslen); - if (rwa->byteswap) { - dmu_object_byteswap_t byteswap = - DMU_OT_BYTESWAP(drro->drr_bonustype); - dmu_ot_byteswap[byteswap].ob_func(db->db_data, - drro->drr_bonuslen); - } - dmu_buf_rele(db, FTAG); - } - dmu_tx_commit(tx); - - return (0); -} - -/* ARGSUSED */ -static int -receive_freeobjects(struct receive_writer_arg *rwa, - struct drr_freeobjects *drrfo) -{ - uint64_t obj; - int next_err = 0; - - if (drrfo->drr_firstobj + drrfo->drr_numobjs < drrfo->drr_firstobj) - return (SET_ERROR(EINVAL)); - - for (obj = drrfo->drr_firstobj == 0 ? 1 : drrfo->drr_firstobj; - obj < drrfo->drr_firstobj + drrfo->drr_numobjs && next_err == 0; - next_err = dmu_object_next(rwa->os, &obj, FALSE, 0)) { - int err; - - err = dmu_object_info(rwa->os, obj, NULL); - if (err == ENOENT) - continue; - else if (err != 0) - return (err); - - err = dmu_free_long_object(rwa->os, obj); - if (err != 0) - return (err); - - if (obj > rwa->max_object) - rwa->max_object = obj; - } - if (next_err != ESRCH) - return (next_err); - return (0); -} - -static int -receive_write(struct receive_writer_arg *rwa, struct drr_write *drrw, - arc_buf_t *abuf) -{ - dmu_tx_t *tx; - int err; - - if (drrw->drr_offset + drrw->drr_logical_size < drrw->drr_offset || - !DMU_OT_IS_VALID(drrw->drr_type)) - return (SET_ERROR(EINVAL)); - - /* - * For resuming to work, records must be in increasing order - * by (object, offset). - */ - if (drrw->drr_object < rwa->last_object || - (drrw->drr_object == rwa->last_object && - drrw->drr_offset < rwa->last_offset)) { - return (SET_ERROR(EINVAL)); - } - rwa->last_object = drrw->drr_object; - rwa->last_offset = drrw->drr_offset; - - if (rwa->last_object > rwa->max_object) - rwa->max_object = rwa->last_object; - - if (dmu_object_info(rwa->os, drrw->drr_object, NULL) != 0) - return (SET_ERROR(EINVAL)); - - tx = dmu_tx_create(rwa->os); - - dmu_tx_hold_write(tx, drrw->drr_object, - drrw->drr_offset, drrw->drr_logical_size); - err = dmu_tx_assign(tx, TXG_WAIT); - if (err != 0) { - dmu_tx_abort(tx); - return (err); - } - if (rwa->byteswap) { - dmu_object_byteswap_t byteswap = - DMU_OT_BYTESWAP(drrw->drr_type); - dmu_ot_byteswap[byteswap].ob_func(abuf->b_data, - DRR_WRITE_PAYLOAD_SIZE(drrw)); - } - - /* use the bonus buf to look up the dnode in dmu_assign_arcbuf */ - dmu_buf_t *bonus; - if (dmu_bonus_hold(rwa->os, drrw->drr_object, FTAG, &bonus) != 0) - return (SET_ERROR(EINVAL)); - dmu_assign_arcbuf(bonus, drrw->drr_offset, abuf, tx); - - /* - * Note: If the receive fails, we want the resume stream to start - * with the same record that we last successfully received (as opposed - * to the next record), so that we can verify that we are - * resuming from the correct location. - */ - save_resume_state(rwa, drrw->drr_object, drrw->drr_offset, tx); - dmu_tx_commit(tx); - dmu_buf_rele(bonus, FTAG); - - return (0); -} - -/* - * Handle a DRR_WRITE_BYREF record. This record is used in dedup'ed - * streams to refer to a copy of the data that is already on the - * system because it came in earlier in the stream. This function - * finds the earlier copy of the data, and uses that copy instead of - * data from the stream to fulfill this write. - */ -static int -receive_write_byref(struct receive_writer_arg *rwa, - struct drr_write_byref *drrwbr) -{ - dmu_tx_t *tx; - int err; - guid_map_entry_t gmesrch; - guid_map_entry_t *gmep; - avl_index_t where; - objset_t *ref_os = NULL; - dmu_buf_t *dbp; - - if (drrwbr->drr_offset + drrwbr->drr_length < drrwbr->drr_offset) - return (SET_ERROR(EINVAL)); - - /* - * If the GUID of the referenced dataset is different from the - * GUID of the target dataset, find the referenced dataset. - */ - if (drrwbr->drr_toguid != drrwbr->drr_refguid) { - gmesrch.guid = drrwbr->drr_refguid; - if ((gmep = avl_find(rwa->guid_to_ds_map, &gmesrch, - &where)) == NULL) { - return (SET_ERROR(EINVAL)); - } - if (dmu_objset_from_ds(gmep->gme_ds, &ref_os)) - return (SET_ERROR(EINVAL)); - } else { - ref_os = rwa->os; - } - - if (drrwbr->drr_object > rwa->max_object) - rwa->max_object = drrwbr->drr_object; - - err = dmu_buf_hold(ref_os, drrwbr->drr_refobject, - drrwbr->drr_refoffset, FTAG, &dbp, DMU_READ_PREFETCH); - if (err != 0) - return (err); - - tx = dmu_tx_create(rwa->os); - - dmu_tx_hold_write(tx, drrwbr->drr_object, - drrwbr->drr_offset, drrwbr->drr_length); - err = dmu_tx_assign(tx, TXG_WAIT); - if (err != 0) { - dmu_tx_abort(tx); - return (err); - } - dmu_write(rwa->os, drrwbr->drr_object, - drrwbr->drr_offset, drrwbr->drr_length, dbp->db_data, tx); - dmu_buf_rele(dbp, FTAG); - - /* See comment in restore_write. */ - save_resume_state(rwa, drrwbr->drr_object, drrwbr->drr_offset, tx); - dmu_tx_commit(tx); - return (0); -} - -static int -receive_write_embedded(struct receive_writer_arg *rwa, - struct drr_write_embedded *drrwe, void *data) -{ - dmu_tx_t *tx; - int err; - - if (drrwe->drr_offset + drrwe->drr_length < drrwe->drr_offset) - return (EINVAL); - - if (drrwe->drr_psize > BPE_PAYLOAD_SIZE) - return (EINVAL); - - if (drrwe->drr_etype >= NUM_BP_EMBEDDED_TYPES) - return (EINVAL); - if (drrwe->drr_compression >= ZIO_COMPRESS_FUNCTIONS) - return (EINVAL); - - if (drrwe->drr_object > rwa->max_object) - rwa->max_object = drrwe->drr_object; - - tx = dmu_tx_create(rwa->os); - - dmu_tx_hold_write(tx, drrwe->drr_object, - drrwe->drr_offset, drrwe->drr_length); - err = dmu_tx_assign(tx, TXG_WAIT); - if (err != 0) { - dmu_tx_abort(tx); - return (err); - } - - dmu_write_embedded(rwa->os, drrwe->drr_object, - drrwe->drr_offset, data, drrwe->drr_etype, - drrwe->drr_compression, drrwe->drr_lsize, drrwe->drr_psize, - rwa->byteswap ^ ZFS_HOST_BYTEORDER, tx); - - /* See comment in restore_write. */ - save_resume_state(rwa, drrwe->drr_object, drrwe->drr_offset, tx); - dmu_tx_commit(tx); - return (0); -} - -static int -receive_spill(struct receive_writer_arg *rwa, struct drr_spill *drrs, - void *data) -{ - dmu_tx_t *tx; - dmu_buf_t *db, *db_spill; - int err; - - if (drrs->drr_length < SPA_MINBLOCKSIZE || - drrs->drr_length > spa_maxblocksize(dmu_objset_spa(rwa->os))) - return (SET_ERROR(EINVAL)); - - if (dmu_object_info(rwa->os, drrs->drr_object, NULL) != 0) - return (SET_ERROR(EINVAL)); - - if (drrs->drr_object > rwa->max_object) - rwa->max_object = drrs->drr_object; - - VERIFY0(dmu_bonus_hold(rwa->os, drrs->drr_object, FTAG, &db)); - if ((err = dmu_spill_hold_by_bonus(db, FTAG, &db_spill)) != 0) { - dmu_buf_rele(db, FTAG); - return (err); - } - - tx = dmu_tx_create(rwa->os); - - dmu_tx_hold_spill(tx, db->db_object); - - err = dmu_tx_assign(tx, TXG_WAIT); - if (err != 0) { - dmu_buf_rele(db, FTAG); - dmu_buf_rele(db_spill, FTAG); - dmu_tx_abort(tx); - return (err); - } - dmu_buf_will_dirty(db_spill, tx); - - if (db_spill->db_size < drrs->drr_length) - VERIFY(0 == dbuf_spill_set_blksz(db_spill, - drrs->drr_length, tx)); - bcopy(data, db_spill->db_data, drrs->drr_length); - - dmu_buf_rele(db, FTAG); - dmu_buf_rele(db_spill, FTAG); - - dmu_tx_commit(tx); - return (0); -} - -/* ARGSUSED */ -static int -receive_free(struct receive_writer_arg *rwa, struct drr_free *drrf) -{ - int err; - - if (drrf->drr_length != -1ULL && - drrf->drr_offset + drrf->drr_length < drrf->drr_offset) - return (SET_ERROR(EINVAL)); - - if (dmu_object_info(rwa->os, drrf->drr_object, NULL) != 0) - return (SET_ERROR(EINVAL)); - - if (drrf->drr_object > rwa->max_object) - rwa->max_object = drrf->drr_object; - - err = dmu_free_long_range(rwa->os, drrf->drr_object, - drrf->drr_offset, drrf->drr_length); - - return (err); -} - -/* used to destroy the drc_ds on error */ -static void -dmu_recv_cleanup_ds(dmu_recv_cookie_t *drc) -{ - if (drc->drc_resumable) { - /* wait for our resume state to be written to disk */ - txg_wait_synced(drc->drc_ds->ds_dir->dd_pool, 0); - dsl_dataset_disown(drc->drc_ds, dmu_recv_tag); - } else { - char name[ZFS_MAX_DATASET_NAME_LEN]; - dsl_dataset_name(drc->drc_ds, name); - dsl_dataset_disown(drc->drc_ds, dmu_recv_tag); - (void) dsl_destroy_head(name); - } -} - -static void -receive_cksum(struct receive_arg *ra, int len, void *buf) -{ - if (ra->byteswap) { - (void) fletcher_4_incremental_byteswap(buf, len, &ra->cksum); - } else { - (void) fletcher_4_incremental_native(buf, len, &ra->cksum); - } -} - -/* - * Read the payload into a buffer of size len, and update the current record's - * payload field. - * Allocate ra->next_rrd and read the next record's header into - * ra->next_rrd->header. - * Verify checksum of payload and next record. - */ -static int -receive_read_payload_and_next_header(struct receive_arg *ra, int len, void *buf) -{ - int err; - - if (len != 0) { - ASSERT3U(len, <=, SPA_MAXBLOCKSIZE); - err = receive_read(ra, len, buf); - if (err != 0) - return (err); - receive_cksum(ra, len, buf); - - /* note: rrd is NULL when reading the begin record's payload */ - if (ra->rrd != NULL) { - ra->rrd->payload = buf; - ra->rrd->payload_size = len; - ra->rrd->bytes_read = ra->bytes_read; - } - } - - ra->prev_cksum = ra->cksum; - - ra->next_rrd = kmem_zalloc(sizeof (*ra->next_rrd), KM_SLEEP); - err = receive_read(ra, sizeof (ra->next_rrd->header), - &ra->next_rrd->header); - ra->next_rrd->bytes_read = ra->bytes_read; - if (err != 0) { - kmem_free(ra->next_rrd, sizeof (*ra->next_rrd)); - ra->next_rrd = NULL; - return (err); - } - if (ra->next_rrd->header.drr_type == DRR_BEGIN) { - kmem_free(ra->next_rrd, sizeof (*ra->next_rrd)); - ra->next_rrd = NULL; - return (SET_ERROR(EINVAL)); - } - - /* - * Note: checksum is of everything up to but not including the - * checksum itself. - */ - ASSERT3U(offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), - ==, sizeof (dmu_replay_record_t) - sizeof (zio_cksum_t)); - receive_cksum(ra, - offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), - &ra->next_rrd->header); - - zio_cksum_t cksum_orig = - ra->next_rrd->header.drr_u.drr_checksum.drr_checksum; - zio_cksum_t *cksump = - &ra->next_rrd->header.drr_u.drr_checksum.drr_checksum; - - if (ra->byteswap) - byteswap_record(&ra->next_rrd->header); - - if ((!ZIO_CHECKSUM_IS_ZERO(cksump)) && - !ZIO_CHECKSUM_EQUAL(ra->cksum, *cksump)) { - kmem_free(ra->next_rrd, sizeof (*ra->next_rrd)); - ra->next_rrd = NULL; - return (SET_ERROR(ECKSUM)); - } - - receive_cksum(ra, sizeof (cksum_orig), &cksum_orig); - - return (0); -} - -static void -objlist_create(struct objlist *list) -{ - list_create(&list->list, sizeof (struct receive_objnode), - offsetof(struct receive_objnode, node)); - list->last_lookup = 0; -} - -static void -objlist_destroy(struct objlist *list) -{ - for (struct receive_objnode *n = list_remove_head(&list->list); - n != NULL; n = list_remove_head(&list->list)) { - kmem_free(n, sizeof (*n)); - } - list_destroy(&list->list); -} - -/* - * This function looks through the objlist to see if the specified object number - * is contained in the objlist. In the process, it will remove all object - * numbers in the list that are smaller than the specified object number. Thus, - * any lookup of an object number smaller than a previously looked up object - * number will always return false; therefore, all lookups should be done in - * ascending order. - */ -static boolean_t -objlist_exists(struct objlist *list, uint64_t object) -{ - struct receive_objnode *node = list_head(&list->list); - ASSERT3U(object, >=, list->last_lookup); - list->last_lookup = object; - while (node != NULL && node->object < object) { - VERIFY3P(node, ==, list_remove_head(&list->list)); - kmem_free(node, sizeof (*node)); - node = list_head(&list->list); - } - return (node != NULL && node->object == object); -} - -/* - * The objlist is a list of object numbers stored in ascending order. However, - * the insertion of new object numbers does not seek out the correct location to - * store a new object number; instead, it appends it to the list for simplicity. - * Thus, any users must take care to only insert new object numbers in ascending - * order. - */ -static void -objlist_insert(struct objlist *list, uint64_t object) -{ - struct receive_objnode *node = kmem_zalloc(sizeof (*node), KM_SLEEP); - node->object = object; -#ifdef ZFS_DEBUG - struct receive_objnode *last_object = list_tail(&list->list); - uint64_t last_objnum = (last_object != NULL ? last_object->object : 0); - ASSERT3U(node->object, >, last_objnum); -#endif - list_insert_tail(&list->list, node); -} - -/* - * Issue the prefetch reads for any necessary indirect blocks. - * - * We use the object ignore list to tell us whether or not to issue prefetches - * for a given object. We do this for both correctness (in case the blocksize - * of an object has changed) and performance (if the object doesn't exist, don't - * needlessly try to issue prefetches). We also trim the list as we go through - * the stream to prevent it from growing to an unbounded size. - * - * The object numbers within will always be in sorted order, and any write - * records we see will also be in sorted order, but they're not sorted with - * respect to each other (i.e. we can get several object records before - * receiving each object's write records). As a result, once we've reached a - * given object number, we can safely remove any reference to lower object - * numbers in the ignore list. In practice, we receive up to 32 object records - * before receiving write records, so the list can have up to 32 nodes in it. - */ -/* ARGSUSED */ -static void -receive_read_prefetch(struct receive_arg *ra, - uint64_t object, uint64_t offset, uint64_t length) -{ - if (!objlist_exists(&ra->ignore_objlist, object)) { - dmu_prefetch(ra->os, object, 1, offset, length, - ZIO_PRIORITY_SYNC_READ); - } -} - -/* - * Read records off the stream, issuing any necessary prefetches. - */ -static int -receive_read_record(struct receive_arg *ra) -{ - int err; - - switch (ra->rrd->header.drr_type) { - case DRR_OBJECT: - { - struct drr_object *drro = &ra->rrd->header.drr_u.drr_object; - uint32_t size = P2ROUNDUP(drro->drr_bonuslen, 8); - void *buf = NULL; - dmu_object_info_t doi; - - if (size > 0) - buf = kmem_zalloc(size, KM_SLEEP); - - err = receive_read_payload_and_next_header(ra, size, buf); - if (err != 0) { - kmem_free(buf, size); - return (err); - } - err = dmu_object_info(ra->os, drro->drr_object, &doi); - /* - * See receive_read_prefetch for an explanation why we're - * storing this object in the ignore_obj_list. - */ - if (err == ENOENT || - (err == 0 && doi.doi_data_block_size != drro->drr_blksz)) { - objlist_insert(&ra->ignore_objlist, drro->drr_object); - err = 0; - } - return (err); - } - case DRR_FREEOBJECTS: - { - err = receive_read_payload_and_next_header(ra, 0, NULL); - return (err); - } - case DRR_WRITE: - { - struct drr_write *drrw = &ra->rrd->header.drr_u.drr_write; - arc_buf_t *abuf; - boolean_t is_meta = DMU_OT_IS_METADATA(drrw->drr_type); - if (DRR_WRITE_COMPRESSED(drrw)) { - ASSERT3U(drrw->drr_compressed_size, >, 0); - ASSERT3U(drrw->drr_logical_size, >=, - drrw->drr_compressed_size); - ASSERT(!is_meta); - abuf = arc_loan_compressed_buf( - dmu_objset_spa(ra->os), - drrw->drr_compressed_size, drrw->drr_logical_size, - drrw->drr_compressiontype); - } else { - abuf = arc_loan_buf(dmu_objset_spa(ra->os), - is_meta, drrw->drr_logical_size); - } - - err = receive_read_payload_and_next_header(ra, - DRR_WRITE_PAYLOAD_SIZE(drrw), abuf->b_data); - if (err != 0) { - dmu_return_arcbuf(abuf); - return (err); - } - ra->rrd->write_buf = abuf; - receive_read_prefetch(ra, drrw->drr_object, drrw->drr_offset, - drrw->drr_logical_size); - return (err); - } - case DRR_WRITE_BYREF: - { - struct drr_write_byref *drrwb = - &ra->rrd->header.drr_u.drr_write_byref; - err = receive_read_payload_and_next_header(ra, 0, NULL); - receive_read_prefetch(ra, drrwb->drr_object, drrwb->drr_offset, - drrwb->drr_length); - return (err); - } - case DRR_WRITE_EMBEDDED: - { - struct drr_write_embedded *drrwe = - &ra->rrd->header.drr_u.drr_write_embedded; - uint32_t size = P2ROUNDUP(drrwe->drr_psize, 8); - void *buf = kmem_zalloc(size, KM_SLEEP); - - err = receive_read_payload_and_next_header(ra, size, buf); - if (err != 0) { - kmem_free(buf, size); - return (err); - } - - receive_read_prefetch(ra, drrwe->drr_object, drrwe->drr_offset, - drrwe->drr_length); - return (err); - } - case DRR_FREE: - { - /* - * It might be beneficial to prefetch indirect blocks here, but - * we don't really have the data to decide for sure. - */ - err = receive_read_payload_and_next_header(ra, 0, NULL); - return (err); - } - case DRR_END: - { - struct drr_end *drre = &ra->rrd->header.drr_u.drr_end; - if (!ZIO_CHECKSUM_EQUAL(ra->prev_cksum, drre->drr_checksum)) - return (SET_ERROR(ECKSUM)); - return (0); - } - case DRR_SPILL: - { - struct drr_spill *drrs = &ra->rrd->header.drr_u.drr_spill; - void *buf = kmem_zalloc(drrs->drr_length, KM_SLEEP); - err = receive_read_payload_and_next_header(ra, drrs->drr_length, - buf); - if (err != 0) - kmem_free(buf, drrs->drr_length); - return (err); - } - default: - return (SET_ERROR(EINVAL)); - } -} - -/* - * Commit the records to the pool. - */ -static int -receive_process_record(struct receive_writer_arg *rwa, - struct receive_record_arg *rrd) -{ - int err; - - /* Processing in order, therefore bytes_read should be increasing. */ - ASSERT3U(rrd->bytes_read, >=, rwa->bytes_read); - rwa->bytes_read = rrd->bytes_read; - - switch (rrd->header.drr_type) { - case DRR_OBJECT: - { - struct drr_object *drro = &rrd->header.drr_u.drr_object; - err = receive_object(rwa, drro, rrd->payload); - kmem_free(rrd->payload, rrd->payload_size); - rrd->payload = NULL; - return (err); - } - case DRR_FREEOBJECTS: - { - struct drr_freeobjects *drrfo = - &rrd->header.drr_u.drr_freeobjects; - return (receive_freeobjects(rwa, drrfo)); - } - case DRR_WRITE: - { - struct drr_write *drrw = &rrd->header.drr_u.drr_write; - err = receive_write(rwa, drrw, rrd->write_buf); - /* if receive_write() is successful, it consumes the arc_buf */ - if (err != 0) - dmu_return_arcbuf(rrd->write_buf); - rrd->write_buf = NULL; - rrd->payload = NULL; - return (err); - } - case DRR_WRITE_BYREF: - { - struct drr_write_byref *drrwbr = - &rrd->header.drr_u.drr_write_byref; - return (receive_write_byref(rwa, drrwbr)); - } - case DRR_WRITE_EMBEDDED: - { - struct drr_write_embedded *drrwe = - &rrd->header.drr_u.drr_write_embedded; - err = receive_write_embedded(rwa, drrwe, rrd->payload); - kmem_free(rrd->payload, rrd->payload_size); - rrd->payload = NULL; - return (err); - } - case DRR_FREE: - { - struct drr_free *drrf = &rrd->header.drr_u.drr_free; - return (receive_free(rwa, drrf)); - } - case DRR_SPILL: - { - struct drr_spill *drrs = &rrd->header.drr_u.drr_spill; - err = receive_spill(rwa, drrs, rrd->payload); - kmem_free(rrd->payload, rrd->payload_size); - rrd->payload = NULL; - return (err); - } - default: - return (SET_ERROR(EINVAL)); - } -} - -/* - * dmu_recv_stream's worker thread; pull records off the queue, and then call - * receive_process_record When we're done, signal the main thread and exit. - */ -static void -receive_writer_thread(void *arg) -{ - struct receive_writer_arg *rwa = arg; - struct receive_record_arg *rrd; - for (rrd = bqueue_dequeue(&rwa->q); !rrd->eos_marker; - rrd = bqueue_dequeue(&rwa->q)) { - /* - * If there's an error, the main thread will stop putting things - * on the queue, but we need to clear everything in it before we - * can exit. - */ - if (rwa->err == 0) { - rwa->err = receive_process_record(rwa, rrd); - } else if (rrd->write_buf != NULL) { - dmu_return_arcbuf(rrd->write_buf); - rrd->write_buf = NULL; - rrd->payload = NULL; - } else if (rrd->payload != NULL) { - kmem_free(rrd->payload, rrd->payload_size); - rrd->payload = NULL; - } - kmem_free(rrd, sizeof (*rrd)); - } - kmem_free(rrd, sizeof (*rrd)); - mutex_enter(&rwa->mutex); - rwa->done = B_TRUE; - cv_signal(&rwa->cv); - mutex_exit(&rwa->mutex); - thread_exit(); -} - -static int -resume_check(struct receive_arg *ra, nvlist_t *begin_nvl) -{ - uint64_t val; - objset_t *mos = dmu_objset_pool(ra->os)->dp_meta_objset; - uint64_t dsobj = dmu_objset_id(ra->os); - uint64_t resume_obj, resume_off; - - if (nvlist_lookup_uint64(begin_nvl, - "resume_object", &resume_obj) != 0 || - nvlist_lookup_uint64(begin_nvl, - "resume_offset", &resume_off) != 0) { - return (SET_ERROR(EINVAL)); - } - VERIFY0(zap_lookup(mos, dsobj, - DS_FIELD_RESUME_OBJECT, sizeof (val), 1, &val)); - if (resume_obj != val) - return (SET_ERROR(EINVAL)); - VERIFY0(zap_lookup(mos, dsobj, - DS_FIELD_RESUME_OFFSET, sizeof (val), 1, &val)); - if (resume_off != val) - return (SET_ERROR(EINVAL)); - - return (0); -} - -/* - * Read in the stream's records, one by one, and apply them to the pool. There - * are two threads involved; the thread that calls this function will spin up a - * worker thread, read the records off the stream one by one, and issue - * prefetches for any necessary indirect blocks. It will then push the records - * onto an internal blocking queue. The worker thread will pull the records off - * the queue, and actually write the data into the DMU. This way, the worker - * thread doesn't have to wait for reads to complete, since everything it needs - * (the indirect blocks) will be prefetched. - * - * NB: callers *must* call dmu_recv_end() if this succeeds. - */ -int -dmu_recv_stream(dmu_recv_cookie_t *drc, vnode_t *vp, offset_t *voffp, - int cleanup_fd, uint64_t *action_handlep) -{ - int err = 0; - struct receive_arg ra = { 0 }; - struct receive_writer_arg rwa = { 0 }; - int featureflags; - nvlist_t *begin_nvl = NULL; - - ra.byteswap = drc->drc_byteswap; - ra.cksum = drc->drc_cksum; - ra.vp = vp; - ra.voff = *voffp; - - if (dsl_dataset_is_zapified(drc->drc_ds)) { - (void) zap_lookup(drc->drc_ds->ds_dir->dd_pool->dp_meta_objset, - drc->drc_ds->ds_object, DS_FIELD_RESUME_BYTES, - sizeof (ra.bytes_read), 1, &ra.bytes_read); - } - - objlist_create(&ra.ignore_objlist); - - /* these were verified in dmu_recv_begin */ - ASSERT3U(DMU_GET_STREAM_HDRTYPE(drc->drc_drrb->drr_versioninfo), ==, - DMU_SUBSTREAM); - ASSERT3U(drc->drc_drrb->drr_type, <, DMU_OST_NUMTYPES); - - /* - * Open the objset we are modifying. - */ - VERIFY0(dmu_objset_from_ds(drc->drc_ds, &ra.os)); - - ASSERT(dsl_dataset_phys(drc->drc_ds)->ds_flags & DS_FLAG_INCONSISTENT); - - featureflags = DMU_GET_FEATUREFLAGS(drc->drc_drrb->drr_versioninfo); - - /* if this stream is dedup'ed, set up the avl tree for guid mapping */ - if (featureflags & DMU_BACKUP_FEATURE_DEDUP) { - minor_t minor; - - if (cleanup_fd == -1) { - ra.err = SET_ERROR(EBADF); - goto out; - } - ra.err = zfs_onexit_fd_hold(cleanup_fd, &minor); - if (ra.err != 0) { - cleanup_fd = -1; - goto out; - } - - if (*action_handlep == 0) { - rwa.guid_to_ds_map = - kmem_alloc(sizeof (avl_tree_t), KM_SLEEP); - avl_create(rwa.guid_to_ds_map, guid_compare, - sizeof (guid_map_entry_t), - offsetof(guid_map_entry_t, avlnode)); - err = zfs_onexit_add_cb(minor, - free_guid_map_onexit, rwa.guid_to_ds_map, - action_handlep); - if (ra.err != 0) - goto out; - } else { - err = zfs_onexit_cb_data(minor, *action_handlep, - (void **)&rwa.guid_to_ds_map); - if (ra.err != 0) - goto out; - } - - drc->drc_guid_to_ds_map = rwa.guid_to_ds_map; - } - - uint32_t payloadlen = drc->drc_drr_begin->drr_payloadlen; - void *payload = NULL; - if (payloadlen != 0) - payload = kmem_alloc(payloadlen, KM_SLEEP); - - err = receive_read_payload_and_next_header(&ra, payloadlen, payload); - if (err != 0) { - if (payloadlen != 0) - kmem_free(payload, payloadlen); - goto out; - } - if (payloadlen != 0) { - err = nvlist_unpack(payload, payloadlen, &begin_nvl, KM_SLEEP); - kmem_free(payload, payloadlen); - if (err != 0) - goto out; - } - - if (featureflags & DMU_BACKUP_FEATURE_RESUMING) { - err = resume_check(&ra, begin_nvl); - if (err != 0) - goto out; - } - - (void) bqueue_init(&rwa.q, zfs_recv_queue_length, - offsetof(struct receive_record_arg, node)); - cv_init(&rwa.cv, NULL, CV_DEFAULT, NULL); - mutex_init(&rwa.mutex, NULL, MUTEX_DEFAULT, NULL); - rwa.os = ra.os; - rwa.byteswap = drc->drc_byteswap; - rwa.resumable = drc->drc_resumable; - - (void) thread_create(NULL, 0, receive_writer_thread, &rwa, 0, curproc, - TS_RUN, minclsyspri); - /* - * We're reading rwa.err without locks, which is safe since we are the - * only reader, and the worker thread is the only writer. It's ok if we - * miss a write for an iteration or two of the loop, since the writer - * thread will keep freeing records we send it until we send it an eos - * marker. - * - * We can leave this loop in 3 ways: First, if rwa.err is - * non-zero. In that case, the writer thread will free the rrd we just - * pushed. Second, if we're interrupted; in that case, either it's the - * first loop and ra.rrd was never allocated, or it's later, and ra.rrd - * has been handed off to the writer thread who will free it. Finally, - * if receive_read_record fails or we're at the end of the stream, then - * we free ra.rrd and exit. - */ - while (rwa.err == 0) { - if (issig(JUSTLOOKING) && issig(FORREAL)) { - err = SET_ERROR(EINTR); - break; - } - - ASSERT3P(ra.rrd, ==, NULL); - ra.rrd = ra.next_rrd; - ra.next_rrd = NULL; - /* Allocates and loads header into ra.next_rrd */ - err = receive_read_record(&ra); - - if (ra.rrd->header.drr_type == DRR_END || err != 0) { - kmem_free(ra.rrd, sizeof (*ra.rrd)); - ra.rrd = NULL; - break; - } - - bqueue_enqueue(&rwa.q, ra.rrd, - sizeof (struct receive_record_arg) + ra.rrd->payload_size); - ra.rrd = NULL; - } - if (ra.next_rrd == NULL) - ra.next_rrd = kmem_zalloc(sizeof (*ra.next_rrd), KM_SLEEP); - ra.next_rrd->eos_marker = B_TRUE; - bqueue_enqueue(&rwa.q, ra.next_rrd, 1); - - mutex_enter(&rwa.mutex); - while (!rwa.done) { - cv_wait(&rwa.cv, &rwa.mutex); - } - mutex_exit(&rwa.mutex); - - /* - * If we are receiving a full stream as a clone, all object IDs which - * are greater than the maximum ID referenced in the stream are - * by definition unused and must be freed. Note that it's possible that - * we've resumed this send and the first record we received was the END - * record. In that case, max_object would be 0, but we shouldn't start - * freeing all objects from there; instead we should start from the - * resumeobj. - */ - if (drc->drc_clone && drc->drc_drrb->drr_fromguid == 0) { - uint64_t obj; - if (nvlist_lookup_uint64(begin_nvl, "resume_object", &obj) != 0) - obj = 0; - if (rwa.max_object > obj) - obj = rwa.max_object; - obj++; - int free_err = 0; - int next_err = 0; - - while (next_err == 0) { - free_err = dmu_free_long_object(rwa.os, obj); - if (free_err != 0 && free_err != ENOENT) - break; - - next_err = dmu_object_next(rwa.os, &obj, FALSE, 0); - } - - if (err == 0) { - if (free_err != 0 && free_err != ENOENT) - err = free_err; - else if (next_err != ESRCH) - err = next_err; - } - } - - cv_destroy(&rwa.cv); - mutex_destroy(&rwa.mutex); - bqueue_destroy(&rwa.q); - if (err == 0) - err = rwa.err; - -out: - nvlist_free(begin_nvl); - if ((featureflags & DMU_BACKUP_FEATURE_DEDUP) && (cleanup_fd != -1)) - zfs_onexit_fd_rele(cleanup_fd); - - if (err != 0) { - /* - * Clean up references. If receive is not resumable, - * destroy what we created, so we don't leave it in - * the inconsistent state. - */ - dmu_recv_cleanup_ds(drc); - } - - *voffp = ra.voff; - objlist_destroy(&ra.ignore_objlist); - return (err); -} - -static int -dmu_recv_end_check(void *arg, dmu_tx_t *tx) -{ - dmu_recv_cookie_t *drc = arg; - dsl_pool_t *dp = dmu_tx_pool(tx); - int error; - - ASSERT3P(drc->drc_ds->ds_owner, ==, dmu_recv_tag); - - if (!drc->drc_newfs) { - dsl_dataset_t *origin_head; - - error = dsl_dataset_hold(dp, drc->drc_tofs, FTAG, &origin_head); - if (error != 0) - return (error); - if (drc->drc_force) { - /* - * We will destroy any snapshots in tofs (i.e. before - * origin_head) that are after the origin (which is - * the snap before drc_ds, because drc_ds can not - * have any snaps of its own). - */ - uint64_t obj; - - obj = dsl_dataset_phys(origin_head)->ds_prev_snap_obj; - while (obj != - dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj) { - dsl_dataset_t *snap; - error = dsl_dataset_hold_obj(dp, obj, FTAG, - &snap); - if (error != 0) - break; - if (snap->ds_dir != origin_head->ds_dir) - error = SET_ERROR(EINVAL); - if (error == 0) { - error = dsl_destroy_snapshot_check_impl( - snap, B_FALSE); - } - obj = dsl_dataset_phys(snap)->ds_prev_snap_obj; - dsl_dataset_rele(snap, FTAG); - if (error != 0) - break; - } - if (error != 0) { - dsl_dataset_rele(origin_head, FTAG); - return (error); - } - } - error = dsl_dataset_clone_swap_check_impl(drc->drc_ds, - origin_head, drc->drc_force, drc->drc_owner, tx); - if (error != 0) { - dsl_dataset_rele(origin_head, FTAG); - return (error); - } - error = dsl_dataset_snapshot_check_impl(origin_head, - drc->drc_tosnap, tx, B_TRUE, 1, drc->drc_cred); - dsl_dataset_rele(origin_head, FTAG); - if (error != 0) - return (error); - - error = dsl_destroy_head_check_impl(drc->drc_ds, 1); - } else { - error = dsl_dataset_snapshot_check_impl(drc->drc_ds, - drc->drc_tosnap, tx, B_TRUE, 1, drc->drc_cred); - } - return (error); -} - -static void -dmu_recv_end_sync(void *arg, dmu_tx_t *tx) -{ - dmu_recv_cookie_t *drc = arg; - dsl_pool_t *dp = dmu_tx_pool(tx); - - spa_history_log_internal_ds(drc->drc_ds, "finish receiving", - tx, "snap=%s", drc->drc_tosnap); - - if (!drc->drc_newfs) { - dsl_dataset_t *origin_head; - - VERIFY0(dsl_dataset_hold(dp, drc->drc_tofs, FTAG, - &origin_head)); - - if (drc->drc_force) { - /* - * Destroy any snapshots of drc_tofs (origin_head) - * after the origin (the snap before drc_ds). - */ - uint64_t obj; - - obj = dsl_dataset_phys(origin_head)->ds_prev_snap_obj; - while (obj != - dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj) { - dsl_dataset_t *snap; - VERIFY0(dsl_dataset_hold_obj(dp, obj, FTAG, - &snap)); - ASSERT3P(snap->ds_dir, ==, origin_head->ds_dir); - obj = dsl_dataset_phys(snap)->ds_prev_snap_obj; - dsl_destroy_snapshot_sync_impl(snap, - B_FALSE, tx); - dsl_dataset_rele(snap, FTAG); - } - } - VERIFY3P(drc->drc_ds->ds_prev, ==, - origin_head->ds_prev); - - dsl_dataset_clone_swap_sync_impl(drc->drc_ds, - origin_head, tx); - dsl_dataset_snapshot_sync_impl(origin_head, - drc->drc_tosnap, tx); - - /* set snapshot's creation time and guid */ - dmu_buf_will_dirty(origin_head->ds_prev->ds_dbuf, tx); - dsl_dataset_phys(origin_head->ds_prev)->ds_creation_time = - drc->drc_drrb->drr_creation_time; - dsl_dataset_phys(origin_head->ds_prev)->ds_guid = - drc->drc_drrb->drr_toguid; - dsl_dataset_phys(origin_head->ds_prev)->ds_flags &= - ~DS_FLAG_INCONSISTENT; - - dmu_buf_will_dirty(origin_head->ds_dbuf, tx); - dsl_dataset_phys(origin_head)->ds_flags &= - ~DS_FLAG_INCONSISTENT; - - drc->drc_newsnapobj = - dsl_dataset_phys(origin_head)->ds_prev_snap_obj; - - dsl_dataset_rele(origin_head, FTAG); - dsl_destroy_head_sync_impl(drc->drc_ds, tx); - - if (drc->drc_owner != NULL) - VERIFY3P(origin_head->ds_owner, ==, drc->drc_owner); - } else { - dsl_dataset_t *ds = drc->drc_ds; - - dsl_dataset_snapshot_sync_impl(ds, drc->drc_tosnap, tx); - - /* set snapshot's creation time and guid */ - dmu_buf_will_dirty(ds->ds_prev->ds_dbuf, tx); - dsl_dataset_phys(ds->ds_prev)->ds_creation_time = - drc->drc_drrb->drr_creation_time; - dsl_dataset_phys(ds->ds_prev)->ds_guid = - drc->drc_drrb->drr_toguid; - dsl_dataset_phys(ds->ds_prev)->ds_flags &= - ~DS_FLAG_INCONSISTENT; - - dmu_buf_will_dirty(ds->ds_dbuf, tx); - dsl_dataset_phys(ds)->ds_flags &= ~DS_FLAG_INCONSISTENT; - if (dsl_dataset_has_resume_receive_state(ds)) { - (void) zap_remove(dp->dp_meta_objset, ds->ds_object, - DS_FIELD_RESUME_FROMGUID, tx); - (void) zap_remove(dp->dp_meta_objset, ds->ds_object, - DS_FIELD_RESUME_OBJECT, tx); - (void) zap_remove(dp->dp_meta_objset, ds->ds_object, - DS_FIELD_RESUME_OFFSET, tx); - (void) zap_remove(dp->dp_meta_objset, ds->ds_object, - DS_FIELD_RESUME_BYTES, tx); - (void) zap_remove(dp->dp_meta_objset, ds->ds_object, - DS_FIELD_RESUME_TOGUID, tx); - (void) zap_remove(dp->dp_meta_objset, ds->ds_object, - DS_FIELD_RESUME_TONAME, tx); - } - drc->drc_newsnapobj = - dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj; - } - /* - * Release the hold from dmu_recv_begin. This must be done before - * we return to open context, so that when we free the dataset's dnode, - * we can evict its bonus buffer. - */ - dsl_dataset_disown(drc->drc_ds, dmu_recv_tag); - drc->drc_ds = NULL; -} - -static int -add_ds_to_guidmap(const char *name, avl_tree_t *guid_map, uint64_t snapobj) -{ - dsl_pool_t *dp; - dsl_dataset_t *snapds; - guid_map_entry_t *gmep; - int err; - - ASSERT(guid_map != NULL); - - err = dsl_pool_hold(name, FTAG, &dp); - if (err != 0) - return (err); - gmep = kmem_alloc(sizeof (*gmep), KM_SLEEP); - err = dsl_dataset_hold_obj(dp, snapobj, gmep, &snapds); - if (err == 0) { - gmep->guid = dsl_dataset_phys(snapds)->ds_guid; - gmep->gme_ds = snapds; - avl_add(guid_map, gmep); - dsl_dataset_long_hold(snapds, gmep); - } else { - kmem_free(gmep, sizeof (*gmep)); - } - - dsl_pool_rele(dp, FTAG); - return (err); -} - -static int dmu_recv_end_modified_blocks = 3; - -static int -dmu_recv_existing_end(dmu_recv_cookie_t *drc) -{ -#ifdef _KERNEL - /* - * We will be destroying the ds; make sure its origin is unmounted if - * necessary. - */ - char name[ZFS_MAX_DATASET_NAME_LEN]; - dsl_dataset_name(drc->drc_ds, name); - zfs_destroy_unmount_origin(name); -#endif - - return (dsl_sync_task(drc->drc_tofs, - dmu_recv_end_check, dmu_recv_end_sync, drc, - dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL)); -} - -static int -dmu_recv_new_end(dmu_recv_cookie_t *drc) -{ - return (dsl_sync_task(drc->drc_tofs, - dmu_recv_end_check, dmu_recv_end_sync, drc, - dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL)); -} - -int -dmu_recv_end(dmu_recv_cookie_t *drc, void *owner) -{ - int error; - - drc->drc_owner = owner; - - if (drc->drc_newfs) - error = dmu_recv_new_end(drc); - else - error = dmu_recv_existing_end(drc); - - if (error != 0) { - dmu_recv_cleanup_ds(drc); - } else if (drc->drc_guid_to_ds_map != NULL) { - (void) add_ds_to_guidmap(drc->drc_tofs, - drc->drc_guid_to_ds_map, - drc->drc_newsnapobj); - } - return (error); -} - -/* - * Return TRUE if this objset is currently being received into. - */ -boolean_t -dmu_objset_is_receiving(objset_t *os) -{ - return (os->os_dsl_dataset != NULL && - os->os_dsl_dataset->ds_owner == dmu_recv_tag); -} diff --git a/usr/src/uts/common/fs/zfs/dsl_dataset.c b/usr/src/uts/common/fs/zfs/dsl_dataset.c index 27873c0d45..caa6855034 100644 --- a/usr/src/uts/common/fs/zfs/dsl_dataset.c +++ b/usr/src/uts/common/fs/zfs/dsl_dataset.c @@ -56,7 +56,7 @@ #include <sys/dsl_destroy.h> #include <sys/dsl_userhold.h> #include <sys/dsl_bookmark.h> -#include <sys/dmu_send.h> +#include <sys/dmu_recv.h> #include <sys/zio_checksum.h> #include <sys/zio_compress.h> #include <zfs_fletcher.h> diff --git a/usr/src/uts/common/fs/zfs/sys/dmu_recv.h b/usr/src/uts/common/fs/zfs/sys/dmu_recv.h new file mode 100644 index 0000000000..56b69e61b1 --- /dev/null +++ b/usr/src/uts/common/fs/zfs/sys/dmu_recv.h @@ -0,0 +1,64 @@ +/* + * 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 (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. + * Copyright (c) 2012, 2014 by Delphix. All rights reserved. + * Copyright 2011 Nexenta Systems, Inc. All rights reserved. + * Copyright (c) 2013, Joyent, Inc. All rights reserved. + * Copyright (c) 2014 Integros [integros.com] + */ + +#ifndef _DMU_RECV_H +#define _DMU_RECV_H + +#include <sys/inttypes.h> +#include <sys/spa.h> + +extern const char *recv_clone_name; + +typedef struct dmu_recv_cookie { + struct dsl_dataset *drc_ds; + struct dmu_replay_record *drc_drr_begin; + struct drr_begin *drc_drrb; + const char *drc_tofs; + const char *drc_tosnap; + boolean_t drc_newfs; + boolean_t drc_byteswap; + boolean_t drc_force; + boolean_t drc_resumable; + boolean_t drc_clone; + struct avl_tree *drc_guid_to_ds_map; + zio_cksum_t drc_cksum; + uint64_t drc_newsnapobj; + void *drc_owner; + cred_t *drc_cred; +} dmu_recv_cookie_t; + +int dmu_recv_begin(char *tofs, char *tosnap, + struct dmu_replay_record *drr_begin, + boolean_t force, boolean_t resumable, char *origin, dmu_recv_cookie_t *drc); +int dmu_recv_stream(dmu_recv_cookie_t *drc, struct vnode *vp, offset_t *voffp, + int cleanup_fd, uint64_t *action_handlep); +int dmu_recv_end(dmu_recv_cookie_t *drc, void *owner); +boolean_t dmu_objset_is_receiving(objset_t *os); + +#endif /* _DMU_RECV_H */ diff --git a/usr/src/uts/common/fs/zfs/sys/dmu_send.h b/usr/src/uts/common/fs/zfs/sys/dmu_send.h index b8403313e9..65d8e99db6 100644 --- a/usr/src/uts/common/fs/zfs/sys/dmu_send.h +++ b/usr/src/uts/common/fs/zfs/sys/dmu_send.h @@ -39,8 +39,6 @@ struct drr_begin; struct avl_tree; struct dmu_replay_record; -extern const char *recv_clone_name; - int dmu_send(const char *tosnap, const char *fromsnap, boolean_t embedok, boolean_t large_block_ok, boolean_t compressok, int outfd, uint64_t resumeobj, uint64_t resumeoff, @@ -53,30 +51,4 @@ int dmu_send_obj(const char *pool, uint64_t tosnap, uint64_t fromsnap, boolean_t embedok, boolean_t large_block_ok, boolean_t compressok, int outfd, struct vnode *vp, offset_t *off); -typedef struct dmu_recv_cookie { - struct dsl_dataset *drc_ds; - struct dmu_replay_record *drc_drr_begin; - struct drr_begin *drc_drrb; - const char *drc_tofs; - const char *drc_tosnap; - boolean_t drc_newfs; - boolean_t drc_byteswap; - boolean_t drc_force; - boolean_t drc_resumable; - boolean_t drc_clone; - struct avl_tree *drc_guid_to_ds_map; - zio_cksum_t drc_cksum; - uint64_t drc_newsnapobj; - void *drc_owner; - cred_t *drc_cred; -} dmu_recv_cookie_t; - -int dmu_recv_begin(char *tofs, char *tosnap, - struct dmu_replay_record *drr_begin, - boolean_t force, boolean_t resumable, char *origin, dmu_recv_cookie_t *drc); -int dmu_recv_stream(dmu_recv_cookie_t *drc, struct vnode *vp, offset_t *voffp, - int cleanup_fd, uint64_t *action_handlep); -int dmu_recv_end(dmu_recv_cookie_t *drc, void *owner); -boolean_t dmu_objset_is_receiving(objset_t *os); - #endif /* _DMU_SEND_H */ diff --git a/usr/src/uts/common/fs/zfs/sys/dsl_bookmark.h b/usr/src/uts/common/fs/zfs/sys/dsl_bookmark.h index 3591986d7b..e477bb231c 100644 --- a/usr/src/uts/common/fs/zfs/sys/dsl_bookmark.h +++ b/usr/src/uts/common/fs/zfs/sys/dsl_bookmark.h @@ -20,6 +20,7 @@ #define _SYS_DSL_BOOKMARK_H #include <sys/zfs_context.h> +#include <sys/dsl_dataset.h> #ifdef __cplusplus extern "C" { diff --git a/usr/src/uts/common/fs/zfs/zfs_ioctl.c b/usr/src/uts/common/fs/zfs/zfs_ioctl.c index 486b7398a4..1b0dc60587 100644 --- a/usr/src/uts/common/fs/zfs/zfs_ioctl.c +++ b/usr/src/uts/common/fs/zfs/zfs_ioctl.c @@ -181,6 +181,7 @@ #include <sys/dsl_scan.h> #include <sharefs/share.h> #include <sys/dmu_objset.h> +#include <sys/dmu_recv.h> #include <sys/dmu_send.h> #include <sys/dsl_destroy.h> #include <sys/dsl_bookmark.h> |