Index: head/cddl/contrib/opensolaris/cmd/zstreamdump/zstreamdump.c =================================================================== --- head/cddl/contrib/opensolaris/cmd/zstreamdump/zstreamdump.c (revision 286586) +++ head/cddl/contrib/opensolaris/cmd/zstreamdump/zstreamdump.c (revision 286587) @@ -1,588 +1,628 @@ /* * 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 2010 Sun Microsystems, Inc. All rights reserved. * Use is subject to license terms. */ /* - * Copyright (c) 2013 by Delphix. All rights reserved. + * Copyright (c) 2013, 2014 by Delphix. All rights reserved. */ #include #include #include #include #include #include +#include #include #include #include /* * If dump mode is enabled, the number of bytes to print per line */ #define BYTES_PER_LINE 16 /* * If dump mode is enabled, the number of bytes to group together, separated * by newlines or spaces */ #define DUMP_GROUPING 4 uint64_t total_write_size = 0; uint64_t total_stream_len = 0; FILE *send_stream = 0; boolean_t do_byteswap = B_FALSE; boolean_t do_cksum = B_TRUE; static void usage(void) { (void) fprintf(stderr, "usage: zstreamdump [-v] [-C] [-d] < file\n"); (void) fprintf(stderr, "\t -v -- verbose\n"); (void) fprintf(stderr, "\t -C -- suppress checksum verification\n"); (void) fprintf(stderr, "\t -d -- dump contents of blocks modified, " "implies verbose\n"); exit(1); } static void * safe_malloc(size_t size) { void *rv = malloc(size); if (rv == NULL) { (void) fprintf(stderr, "ERROR; failed to allocate %zu bytes\n", size); abort(); } return (rv); } /* * ssread - send stream read. * * Read while computing incremental checksum */ - static size_t ssread(void *buf, size_t len, zio_cksum_t *cksum) { size_t outlen; if ((outlen = fread(buf, len, 1, send_stream)) == 0) return (0); - if (do_cksum && cksum) { + if (do_cksum) { if (do_byteswap) fletcher_4_incremental_byteswap(buf, len, cksum); else fletcher_4_incremental_native(buf, len, cksum); } total_stream_len += len; return (outlen); } +static size_t +read_hdr(dmu_replay_record_t *drr, zio_cksum_t *cksum) +{ + ASSERT3U(offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), + ==, sizeof (dmu_replay_record_t) - sizeof (zio_cksum_t)); + size_t r = ssread(drr, sizeof (*drr) - sizeof (zio_cksum_t), cksum); + if (r == 0) + return (0); + zio_cksum_t saved_cksum = *cksum; + r = ssread(&drr->drr_u.drr_checksum.drr_checksum, + sizeof (zio_cksum_t), cksum); + if (r == 0) + return (0); + if (!ZIO_CHECKSUM_IS_ZERO(&drr->drr_u.drr_checksum.drr_checksum) && + !ZIO_CHECKSUM_EQUAL(saved_cksum, + drr->drr_u.drr_checksum.drr_checksum)) { + fprintf(stderr, "invalid checksum\n"); + (void) printf("Incorrect checksum in record header.\n"); + (void) printf("Expected checksum = %llx/%llx/%llx/%llx\n", + saved_cksum.zc_word[0], + saved_cksum.zc_word[1], + saved_cksum.zc_word[2], + saved_cksum.zc_word[3]); + exit(1); + } + return (sizeof (*drr)); +} + /* * Print part of a block in ASCII characters */ static void print_ascii_block(char *subbuf, int length) { int i; for (i = 0; i < length; i++) { char char_print = isprint(subbuf[i]) ? subbuf[i] : '.'; if (i != 0 && i % DUMP_GROUPING == 0) { (void) printf(" "); } (void) printf("%c", char_print); } (void) printf("\n"); } /* * print_block - Dump the contents of a modified block to STDOUT * * Assume that buf has capacity evenly divisible by BYTES_PER_LINE */ static void print_block(char *buf, int length) { int i; /* * Start printing ASCII characters at a constant offset, after * the hex prints. Leave 3 characters per byte on a line (2 digit * hex number plus 1 space) plus spaces between characters and * groupings. */ int ascii_start = BYTES_PER_LINE * 3 + BYTES_PER_LINE / DUMP_GROUPING + 2; for (i = 0; i < length; i += BYTES_PER_LINE) { int j; int this_line_length = MIN(BYTES_PER_LINE, length - i); int print_offset = 0; for (j = 0; j < this_line_length; j++) { int buf_offset = i + j; /* * Separate every DUMP_GROUPING bytes by a space. */ if (buf_offset % DUMP_GROUPING == 0) { print_offset += printf(" "); } /* * Print the two-digit hex value for this byte. */ unsigned char hex_print = buf[buf_offset]; print_offset += printf("%02x ", hex_print); } (void) printf("%*s", ascii_start - print_offset, " "); print_ascii_block(buf + i, this_line_length); } } int main(int argc, char *argv[]) { char *buf = safe_malloc(SPA_MAXBLOCKSIZE); uint64_t drr_record_count[DRR_NUMTYPES] = { 0 }; uint64_t total_records = 0; dmu_replay_record_t thedrr; dmu_replay_record_t *drr = &thedrr; struct drr_begin *drrb = &thedrr.drr_u.drr_begin; struct drr_end *drre = &thedrr.drr_u.drr_end; struct drr_object *drro = &thedrr.drr_u.drr_object; struct drr_freeobjects *drrfo = &thedrr.drr_u.drr_freeobjects; struct drr_write *drrw = &thedrr.drr_u.drr_write; struct drr_write_byref *drrwbr = &thedrr.drr_u.drr_write_byref; struct drr_free *drrf = &thedrr.drr_u.drr_free; struct drr_spill *drrs = &thedrr.drr_u.drr_spill; struct drr_write_embedded *drrwe = &thedrr.drr_u.drr_write_embedded; + struct drr_checksum *drrc = &thedrr.drr_u.drr_checksum; char c; boolean_t verbose = B_FALSE; + boolean_t very_verbose = B_FALSE; boolean_t first = B_TRUE; /* * dump flag controls whether the contents of any modified data blocks * are printed to the console during processing of the stream. Warning: * for large streams, this can obviously lead to massive prints. */ boolean_t dump = B_FALSE; int err; zio_cksum_t zc = { 0 }; zio_cksum_t pcksum = { 0 }; while ((c = getopt(argc, argv, ":vCd")) != -1) { switch (c) { case 'C': do_cksum = B_FALSE; break; case 'v': + if (verbose) + very_verbose = B_TRUE; verbose = B_TRUE; break; case 'd': dump = B_TRUE; verbose = B_TRUE; + very_verbose = B_TRUE; break; case ':': (void) fprintf(stderr, "missing argument for '%c' option\n", optopt); usage(); break; case '?': (void) fprintf(stderr, "invalid option '%c'\n", optopt); usage(); } } if (isatty(STDIN_FILENO)) { (void) fprintf(stderr, "Error: Backup stream can not be read " "from a terminal.\n" "You must redirect standard input.\n"); exit(1); } send_stream = stdin; pcksum = zc; - while (ssread(drr, sizeof (dmu_replay_record_t), &zc)) { + while (read_hdr(drr, &zc)) { /* * If this is the first DMU record being processed, check for * the magic bytes and figure out the endian-ness based on them. */ if (first) { if (drrb->drr_magic == BSWAP_64(DMU_BACKUP_MAGIC)) { do_byteswap = B_TRUE; if (do_cksum) { ZIO_SET_CHECKSUM(&zc, 0, 0, 0, 0); /* * recalculate header checksum now * that we know it needs to be * byteswapped. */ fletcher_4_incremental_byteswap(drr, sizeof (dmu_replay_record_t), &zc); } } else if (drrb->drr_magic != DMU_BACKUP_MAGIC) { (void) fprintf(stderr, "Invalid stream " "(bad magic number)\n"); exit(1); } first = B_FALSE; } if (do_byteswap) { drr->drr_type = BSWAP_32(drr->drr_type); drr->drr_payloadlen = BSWAP_32(drr->drr_payloadlen); } /* * At this point, the leading fields of the replay record * (drr_type and drr_payloadlen) have been byte-swapped if * necessary, but the rest of the data structure (the * union of type-specific structures) is still in its * original state. */ if (drr->drr_type >= DRR_NUMTYPES) { (void) printf("INVALID record found: type 0x%x\n", drr->drr_type); (void) printf("Aborting.\n"); exit(1); } drr_record_count[drr->drr_type]++; total_records++; switch (drr->drr_type) { case DRR_BEGIN: if (do_byteswap) { drrb->drr_magic = BSWAP_64(drrb->drr_magic); drrb->drr_versioninfo = BSWAP_64(drrb->drr_versioninfo); drrb->drr_creation_time = BSWAP_64(drrb->drr_creation_time); drrb->drr_type = BSWAP_32(drrb->drr_type); drrb->drr_flags = BSWAP_32(drrb->drr_flags); drrb->drr_toguid = BSWAP_64(drrb->drr_toguid); drrb->drr_fromguid = BSWAP_64(drrb->drr_fromguid); } (void) printf("BEGIN record\n"); (void) printf("\thdrtype = %lld\n", DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo)); (void) printf("\tfeatures = %llx\n", DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo)); (void) printf("\tmagic = %llx\n", (u_longlong_t)drrb->drr_magic); (void) printf("\tcreation_time = %llx\n", (u_longlong_t)drrb->drr_creation_time); (void) printf("\ttype = %u\n", drrb->drr_type); (void) printf("\tflags = 0x%x\n", drrb->drr_flags); (void) printf("\ttoguid = %llx\n", (u_longlong_t)drrb->drr_toguid); (void) printf("\tfromguid = %llx\n", (u_longlong_t)drrb->drr_fromguid); (void) printf("\ttoname = %s\n", drrb->drr_toname); if (verbose) (void) printf("\n"); if ((DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == DMU_COMPOUNDSTREAM) && drr->drr_payloadlen != 0) { nvlist_t *nv; int sz = drr->drr_payloadlen; if (sz > SPA_MAXBLOCKSIZE) { free(buf); buf = safe_malloc(sz); } (void) ssread(buf, sz, &zc); if (ferror(send_stream)) perror("fread"); err = nvlist_unpack(buf, sz, &nv, 0); if (err) perror(strerror(err)); nvlist_print(stdout, nv); nvlist_free(nv); } break; case DRR_END: if (do_byteswap) { drre->drr_checksum.zc_word[0] = BSWAP_64(drre->drr_checksum.zc_word[0]); drre->drr_checksum.zc_word[1] = BSWAP_64(drre->drr_checksum.zc_word[1]); drre->drr_checksum.zc_word[2] = BSWAP_64(drre->drr_checksum.zc_word[2]); drre->drr_checksum.zc_word[3] = BSWAP_64(drre->drr_checksum.zc_word[3]); } /* * We compare against the *previous* checksum * value, because the stored checksum is of * everything before the DRR_END record. */ if (do_cksum && !ZIO_CHECKSUM_EQUAL(drre->drr_checksum, pcksum)) { (void) printf("Expected checksum differs from " "checksum in stream.\n"); (void) printf("Expected checksum = " "%llx/%llx/%llx/%llx\n", pcksum.zc_word[0], pcksum.zc_word[1], pcksum.zc_word[2], pcksum.zc_word[3]); } (void) printf("END checksum = %llx/%llx/%llx/%llx\n", drre->drr_checksum.zc_word[0], drre->drr_checksum.zc_word[1], drre->drr_checksum.zc_word[2], drre->drr_checksum.zc_word[3]); ZIO_SET_CHECKSUM(&zc, 0, 0, 0, 0); break; case DRR_OBJECT: if (do_byteswap) { drro->drr_object = BSWAP_64(drro->drr_object); drro->drr_type = BSWAP_32(drro->drr_type); drro->drr_bonustype = BSWAP_32(drro->drr_bonustype); drro->drr_blksz = BSWAP_32(drro->drr_blksz); drro->drr_bonuslen = BSWAP_32(drro->drr_bonuslen); drro->drr_toguid = BSWAP_64(drro->drr_toguid); } if (verbose) { (void) printf("OBJECT object = %llu type = %u " "bonustype = %u blksz = %u bonuslen = %u\n", (u_longlong_t)drro->drr_object, drro->drr_type, drro->drr_bonustype, drro->drr_blksz, drro->drr_bonuslen); } if (drro->drr_bonuslen > 0) { (void) ssread(buf, P2ROUNDUP(drro->drr_bonuslen, 8), &zc); if (dump) { print_block(buf, P2ROUNDUP(drro->drr_bonuslen, 8)); } } break; case DRR_FREEOBJECTS: if (do_byteswap) { drrfo->drr_firstobj = BSWAP_64(drrfo->drr_firstobj); drrfo->drr_numobjs = BSWAP_64(drrfo->drr_numobjs); drrfo->drr_toguid = BSWAP_64(drrfo->drr_toguid); } if (verbose) { (void) printf("FREEOBJECTS firstobj = %llu " "numobjs = %llu\n", (u_longlong_t)drrfo->drr_firstobj, (u_longlong_t)drrfo->drr_numobjs); } break; case DRR_WRITE: if (do_byteswap) { drrw->drr_object = BSWAP_64(drrw->drr_object); drrw->drr_type = BSWAP_32(drrw->drr_type); drrw->drr_offset = BSWAP_64(drrw->drr_offset); drrw->drr_length = BSWAP_64(drrw->drr_length); drrw->drr_toguid = BSWAP_64(drrw->drr_toguid); drrw->drr_key.ddk_prop = BSWAP_64(drrw->drr_key.ddk_prop); } /* * If this is verbose and/or dump output, * print info on the modified block */ if (verbose) { (void) printf("WRITE object = %llu type = %u " "checksum type = %u\n" - "offset = %llu length = %llu " + " offset = %llu length = %llu " "props = %llx\n", (u_longlong_t)drrw->drr_object, drrw->drr_type, drrw->drr_checksumtype, (u_longlong_t)drrw->drr_offset, (u_longlong_t)drrw->drr_length, (u_longlong_t)drrw->drr_key.ddk_prop); } /* * Read the contents of the block in from STDIN to buf */ (void) ssread(buf, drrw->drr_length, &zc); /* * If in dump mode */ if (dump) { print_block(buf, drrw->drr_length); } total_write_size += drrw->drr_length; break; case DRR_WRITE_BYREF: if (do_byteswap) { drrwbr->drr_object = BSWAP_64(drrwbr->drr_object); drrwbr->drr_offset = BSWAP_64(drrwbr->drr_offset); drrwbr->drr_length = BSWAP_64(drrwbr->drr_length); drrwbr->drr_toguid = BSWAP_64(drrwbr->drr_toguid); drrwbr->drr_refguid = BSWAP_64(drrwbr->drr_refguid); drrwbr->drr_refobject = BSWAP_64(drrwbr->drr_refobject); drrwbr->drr_refoffset = BSWAP_64(drrwbr->drr_refoffset); drrwbr->drr_key.ddk_prop = BSWAP_64(drrwbr->drr_key.ddk_prop); } if (verbose) { (void) printf("WRITE_BYREF object = %llu " "checksum type = %u props = %llx\n" - "offset = %llu length = %llu\n" + " offset = %llu length = %llu\n" "toguid = %llx refguid = %llx\n" - "refobject = %llu refoffset = %llu\n", + " refobject = %llu refoffset = %llu\n", (u_longlong_t)drrwbr->drr_object, drrwbr->drr_checksumtype, (u_longlong_t)drrwbr->drr_key.ddk_prop, (u_longlong_t)drrwbr->drr_offset, (u_longlong_t)drrwbr->drr_length, (u_longlong_t)drrwbr->drr_toguid, (u_longlong_t)drrwbr->drr_refguid, (u_longlong_t)drrwbr->drr_refobject, (u_longlong_t)drrwbr->drr_refoffset); } break; case DRR_FREE: if (do_byteswap) { drrf->drr_object = BSWAP_64(drrf->drr_object); drrf->drr_offset = BSWAP_64(drrf->drr_offset); drrf->drr_length = BSWAP_64(drrf->drr_length); } if (verbose) { (void) printf("FREE object = %llu " "offset = %llu length = %lld\n", (u_longlong_t)drrf->drr_object, (u_longlong_t)drrf->drr_offset, (longlong_t)drrf->drr_length); } break; case DRR_SPILL: if (do_byteswap) { drrs->drr_object = BSWAP_64(drrs->drr_object); drrs->drr_length = BSWAP_64(drrs->drr_length); } if (verbose) { (void) printf("SPILL block for object = %llu " "length = %llu\n", drrs->drr_object, drrs->drr_length); } (void) ssread(buf, drrs->drr_length, &zc); if (dump) { print_block(buf, drrs->drr_length); } break; case DRR_WRITE_EMBEDDED: if (do_byteswap) { drrwe->drr_object = BSWAP_64(drrwe->drr_object); drrwe->drr_offset = BSWAP_64(drrwe->drr_offset); drrwe->drr_length = BSWAP_64(drrwe->drr_length); drrwe->drr_toguid = BSWAP_64(drrwe->drr_toguid); drrwe->drr_lsize = BSWAP_32(drrwe->drr_lsize); drrwe->drr_psize = BSWAP_32(drrwe->drr_psize); } if (verbose) { (void) printf("WRITE_EMBEDDED object = %llu " "offset = %llu length = %llu\n" - "toguid = %llx comp = %u etype = %u " + " toguid = %llx comp = %u etype = %u " "lsize = %u psize = %u\n", (u_longlong_t)drrwe->drr_object, (u_longlong_t)drrwe->drr_offset, (u_longlong_t)drrwe->drr_length, (u_longlong_t)drrwe->drr_toguid, drrwe->drr_compression, drrwe->drr_etype, drrwe->drr_lsize, drrwe->drr_psize); } (void) ssread(buf, P2ROUNDUP(drrwe->drr_psize, 8), &zc); break; + } + if (drr->drr_type != DRR_BEGIN && very_verbose) { + (void) printf(" checksum = %llx/%llx/%llx/%llx\n", + (longlong_t)drrc->drr_checksum.zc_word[0], + (longlong_t)drrc->drr_checksum.zc_word[1], + (longlong_t)drrc->drr_checksum.zc_word[2], + (longlong_t)drrc->drr_checksum.zc_word[3]); } pcksum = zc; } free(buf); /* Print final summary */ (void) printf("SUMMARY:\n"); (void) printf("\tTotal DRR_BEGIN records = %lld\n", (u_longlong_t)drr_record_count[DRR_BEGIN]); (void) printf("\tTotal DRR_END records = %lld\n", (u_longlong_t)drr_record_count[DRR_END]); (void) printf("\tTotal DRR_OBJECT records = %lld\n", (u_longlong_t)drr_record_count[DRR_OBJECT]); (void) printf("\tTotal DRR_FREEOBJECTS records = %lld\n", (u_longlong_t)drr_record_count[DRR_FREEOBJECTS]); (void) printf("\tTotal DRR_WRITE records = %lld\n", (u_longlong_t)drr_record_count[DRR_WRITE]); (void) printf("\tTotal DRR_WRITE_BYREF records = %lld\n", (u_longlong_t)drr_record_count[DRR_WRITE_BYREF]); (void) printf("\tTotal DRR_WRITE_EMBEDDED records = %lld\n", (u_longlong_t)drr_record_count[DRR_WRITE_EMBEDDED]); (void) printf("\tTotal DRR_FREE records = %lld\n", (u_longlong_t)drr_record_count[DRR_FREE]); (void) printf("\tTotal DRR_SPILL records = %lld\n", (u_longlong_t)drr_record_count[DRR_SPILL]); (void) printf("\tTotal records = %lld\n", (u_longlong_t)total_records); (void) printf("\tTotal write size = %lld (0x%llx)\n", (u_longlong_t)total_write_size, (u_longlong_t)total_write_size); (void) printf("\tTotal stream length = %lld (0x%llx)\n", (u_longlong_t)total_stream_len, (u_longlong_t)total_stream_len); return (0); } Index: head/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_sendrecv.c =================================================================== --- head/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_sendrecv.c (revision 286586) +++ head/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_sendrecv.c (revision 286587) @@ -1,3334 +1,3331 @@ /* * 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) 2011, 2014 by Delphix. All rights reserved. * Copyright (c) 2012, Joyent, Inc. All rights reserved. * Copyright (c) 2012 Pawel Jakub Dawidek . * All rights reserved. * Copyright (c) 2013 Steven Hartland. All rights reserved. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "zfs_namecheck.h" #include "zfs_prop.h" #include "zfs_fletcher.h" #include "libzfs_impl.h" #include #include #include #ifdef __FreeBSD__ extern int zfs_ioctl_version; #endif /* in libzfs_dataset.c */ extern void zfs_setprop_error(libzfs_handle_t *, zfs_prop_t, int, char *); /* We need to use something for ENODATA. */ #define ENODATA EIDRM static int zfs_receive_impl(libzfs_handle_t *, const char *, recvflags_t *, int, const char *, nvlist_t *, avl_tree_t *, char **, int, uint64_t *); static const zio_cksum_t zero_cksum = { 0 }; typedef struct dedup_arg { int inputfd; int outputfd; libzfs_handle_t *dedup_hdl; } dedup_arg_t; typedef struct progress_arg { zfs_handle_t *pa_zhp; int pa_fd; boolean_t pa_parsable; } progress_arg_t; typedef struct dataref { uint64_t ref_guid; uint64_t ref_object; uint64_t ref_offset; } dataref_t; typedef struct dedup_entry { struct dedup_entry *dde_next; zio_cksum_t dde_chksum; uint64_t dde_prop; dataref_t dde_ref; } dedup_entry_t; #define MAX_DDT_PHYSMEM_PERCENT 20 #define SMALLEST_POSSIBLE_MAX_DDT_MB 128 typedef struct dedup_table { dedup_entry_t **dedup_hash_array; umem_cache_t *ddecache; uint64_t max_ddt_size; /* max dedup table size in bytes */ uint64_t cur_ddt_size; /* current dedup table size in bytes */ uint64_t ddt_count; int numhashbits; boolean_t ddt_full; } dedup_table_t; static int high_order_bit(uint64_t n) { int count; for (count = 0; n != 0; count++) n >>= 1; return (count); } static size_t ssread(void *buf, size_t len, FILE *stream) { size_t outlen; if ((outlen = fread(buf, len, 1, stream)) == 0) return (0); return (outlen); } static void ddt_hash_append(libzfs_handle_t *hdl, dedup_table_t *ddt, dedup_entry_t **ddepp, zio_cksum_t *cs, uint64_t prop, dataref_t *dr) { dedup_entry_t *dde; if (ddt->cur_ddt_size >= ddt->max_ddt_size) { if (ddt->ddt_full == B_FALSE) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "Dedup table full. Deduplication will continue " "with existing table entries")); ddt->ddt_full = B_TRUE; } return; } if ((dde = umem_cache_alloc(ddt->ddecache, UMEM_DEFAULT)) != NULL) { assert(*ddepp == NULL); dde->dde_next = NULL; dde->dde_chksum = *cs; dde->dde_prop = prop; dde->dde_ref = *dr; *ddepp = dde; ddt->cur_ddt_size += sizeof (dedup_entry_t); ddt->ddt_count++; } } /* * Using the specified dedup table, do a lookup for an entry with * the checksum cs. If found, return the block's reference info * in *dr. Otherwise, insert a new entry in the dedup table, using * the reference information specified by *dr. * * return value: true - entry was found * false - entry was not found */ static boolean_t ddt_update(libzfs_handle_t *hdl, dedup_table_t *ddt, zio_cksum_t *cs, uint64_t prop, dataref_t *dr) { uint32_t hashcode; dedup_entry_t **ddepp; hashcode = BF64_GET(cs->zc_word[0], 0, ddt->numhashbits); for (ddepp = &(ddt->dedup_hash_array[hashcode]); *ddepp != NULL; ddepp = &((*ddepp)->dde_next)) { if (ZIO_CHECKSUM_EQUAL(((*ddepp)->dde_chksum), *cs) && (*ddepp)->dde_prop == prop) { *dr = (*ddepp)->dde_ref; return (B_TRUE); } } ddt_hash_append(hdl, ddt, ddepp, cs, prop, dr); return (B_FALSE); } static int -cksum_and_write(const void *buf, uint64_t len, zio_cksum_t *zc, int outfd) +dump_record(dmu_replay_record_t *drr, void *payload, int payload_len, + zio_cksum_t *zc, int outfd) { - fletcher_4_incremental_native(buf, len, zc); - return (write(outfd, buf, len)); + ASSERT3U(offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), + ==, sizeof (dmu_replay_record_t) - sizeof (zio_cksum_t)); + fletcher_4_incremental_native(drr, + offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), zc); + if (drr->drr_type != DRR_BEGIN) { + ASSERT(ZIO_CHECKSUM_IS_ZERO(&drr->drr_u. + drr_checksum.drr_checksum)); + drr->drr_u.drr_checksum.drr_checksum = *zc; + } + fletcher_4_incremental_native(&drr->drr_u.drr_checksum.drr_checksum, + sizeof (zio_cksum_t), zc); + if (write(outfd, drr, sizeof (*drr)) == -1) + return (errno); + if (payload_len != 0) { + fletcher_4_incremental_native(payload, payload_len, zc); + if (write(outfd, payload, payload_len) == -1) + return (errno); + } + return (0); } /* * This function is started in a separate thread when the dedup option * has been requested. The main send thread determines the list of * snapshots to be included in the send stream and makes the ioctl calls * for each one. But instead of having the ioctl send the output to the * the output fd specified by the caller of zfs_send()), the * ioctl is told to direct the output to a pipe, which is read by the * alternate thread running THIS function. This function does the * dedup'ing by: * 1. building a dedup table (the DDT) * 2. doing checksums on each data block and inserting a record in the DDT * 3. looking for matching checksums, and * 4. sending a DRR_WRITE_BYREF record instead of a write record whenever * a duplicate block is found. * The output of this function then goes to the output fd requested * by the caller of zfs_send(). */ static void * cksummer(void *arg) { dedup_arg_t *dda = arg; char *buf = zfs_alloc(dda->dedup_hdl, SPA_MAXBLOCKSIZE); dmu_replay_record_t thedrr; dmu_replay_record_t *drr = &thedrr; - struct drr_begin *drrb = &thedrr.drr_u.drr_begin; - struct drr_end *drre = &thedrr.drr_u.drr_end; - struct drr_object *drro = &thedrr.drr_u.drr_object; - struct drr_write *drrw = &thedrr.drr_u.drr_write; - struct drr_spill *drrs = &thedrr.drr_u.drr_spill; - struct drr_write_embedded *drrwe = &thedrr.drr_u.drr_write_embedded; FILE *ofp; int outfd; - dmu_replay_record_t wbr_drr = {0}; - struct drr_write_byref *wbr_drrr = &wbr_drr.drr_u.drr_write_byref; dedup_table_t ddt; zio_cksum_t stream_cksum; uint64_t physmem = sysconf(_SC_PHYS_PAGES) * sysconf(_SC_PAGESIZE); uint64_t numbuckets; ddt.max_ddt_size = - MAX((physmem * MAX_DDT_PHYSMEM_PERCENT)/100, - SMALLEST_POSSIBLE_MAX_DDT_MB<<20); + MAX((physmem * MAX_DDT_PHYSMEM_PERCENT) / 100, + SMALLEST_POSSIBLE_MAX_DDT_MB << 20); - numbuckets = ddt.max_ddt_size/(sizeof (dedup_entry_t)); + numbuckets = ddt.max_ddt_size / (sizeof (dedup_entry_t)); /* * numbuckets must be a power of 2. Increase number to * a power of 2 if necessary. */ if (!ISP2(numbuckets)) numbuckets = 1 << high_order_bit(numbuckets); ddt.dedup_hash_array = calloc(numbuckets, sizeof (dedup_entry_t *)); ddt.ddecache = umem_cache_create("dde", sizeof (dedup_entry_t), 0, NULL, NULL, NULL, NULL, NULL, 0); ddt.cur_ddt_size = numbuckets * sizeof (dedup_entry_t *); ddt.numhashbits = high_order_bit(numbuckets) - 1; ddt.ddt_full = B_FALSE; - /* Initialize the write-by-reference block. */ - wbr_drr.drr_type = DRR_WRITE_BYREF; - wbr_drr.drr_payloadlen = 0; - outfd = dda->outputfd; ofp = fdopen(dda->inputfd, "r"); - while (ssread(drr, sizeof (dmu_replay_record_t), ofp) != 0) { + while (ssread(drr, sizeof (*drr), ofp) != 0) { switch (drr->drr_type) { case DRR_BEGIN: { - int fflags; + struct drr_begin *drrb = &drr->drr_u.drr_begin; + int fflags; + int sz = 0; ZIO_SET_CHECKSUM(&stream_cksum, 0, 0, 0, 0); + ASSERT3U(drrb->drr_magic, ==, DMU_BACKUP_MAGIC); + /* set the DEDUP feature flag for this stream */ fflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo); fflags |= (DMU_BACKUP_FEATURE_DEDUP | DMU_BACKUP_FEATURE_DEDUPPROPS); DMU_SET_FEATUREFLAGS(drrb->drr_versioninfo, fflags); - if (cksum_and_write(drr, sizeof (dmu_replay_record_t), - &stream_cksum, outfd) == -1) - goto out; if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == DMU_COMPOUNDSTREAM && drr->drr_payloadlen != 0) { - int sz = drr->drr_payloadlen; + sz = drr->drr_payloadlen; if (sz > SPA_MAXBLOCKSIZE) { buf = zfs_realloc(dda->dedup_hdl, buf, SPA_MAXBLOCKSIZE, sz); } (void) ssread(buf, sz, ofp); if (ferror(stdin)) perror("fread"); - if (cksum_and_write(buf, sz, &stream_cksum, - outfd) == -1) - goto out; } + if (dump_record(drr, buf, sz, &stream_cksum, + outfd) != 0) + goto out; break; } case DRR_END: { + struct drr_end *drre = &drr->drr_u.drr_end; /* use the recalculated checksum */ - ZIO_SET_CHECKSUM(&drre->drr_checksum, - stream_cksum.zc_word[0], stream_cksum.zc_word[1], - stream_cksum.zc_word[2], stream_cksum.zc_word[3]); - if ((write(outfd, drr, - sizeof (dmu_replay_record_t))) == -1) + drre->drr_checksum = stream_cksum; + if (dump_record(drr, NULL, 0, &stream_cksum, + outfd) != 0) goto out; break; } case DRR_OBJECT: { - if (cksum_and_write(drr, sizeof (dmu_replay_record_t), - &stream_cksum, outfd) == -1) - goto out; + struct drr_object *drro = &drr->drr_u.drr_object; if (drro->drr_bonuslen > 0) { (void) ssread(buf, P2ROUNDUP((uint64_t)drro->drr_bonuslen, 8), ofp); - if (cksum_and_write(buf, - P2ROUNDUP((uint64_t)drro->drr_bonuslen, 8), - &stream_cksum, outfd) == -1) - goto out; } + if (dump_record(drr, buf, + P2ROUNDUP((uint64_t)drro->drr_bonuslen, 8), + &stream_cksum, outfd) != 0) + goto out; break; } case DRR_SPILL: { - if (cksum_and_write(drr, sizeof (dmu_replay_record_t), - &stream_cksum, outfd) == -1) - goto out; + struct drr_spill *drrs = &drr->drr_u.drr_spill; (void) ssread(buf, drrs->drr_length, ofp); - if (cksum_and_write(buf, drrs->drr_length, - &stream_cksum, outfd) == -1) + if (dump_record(drr, buf, drrs->drr_length, + &stream_cksum, outfd) != 0) goto out; break; } case DRR_FREEOBJECTS: { - if (cksum_and_write(drr, sizeof (dmu_replay_record_t), - &stream_cksum, outfd) == -1) + if (dump_record(drr, NULL, 0, &stream_cksum, + outfd) != 0) goto out; break; } case DRR_WRITE: { + struct drr_write *drrw = &drr->drr_u.drr_write; dataref_t dataref; (void) ssread(buf, drrw->drr_length, ofp); /* * Use the existing checksum if it's dedup-capable, * else calculate a SHA256 checksum for it. */ if (ZIO_CHECKSUM_EQUAL(drrw->drr_key.ddk_cksum, zero_cksum) || !DRR_IS_DEDUP_CAPABLE(drrw->drr_checksumflags)) { SHA256_CTX ctx; zio_cksum_t tmpsha256; SHA256Init(&ctx); SHA256Update(&ctx, buf, drrw->drr_length); SHA256Final(&tmpsha256, &ctx); drrw->drr_key.ddk_cksum.zc_word[0] = BE_64(tmpsha256.zc_word[0]); drrw->drr_key.ddk_cksum.zc_word[1] = BE_64(tmpsha256.zc_word[1]); drrw->drr_key.ddk_cksum.zc_word[2] = BE_64(tmpsha256.zc_word[2]); drrw->drr_key.ddk_cksum.zc_word[3] = BE_64(tmpsha256.zc_word[3]); drrw->drr_checksumtype = ZIO_CHECKSUM_SHA256; drrw->drr_checksumflags = DRR_CHECKSUM_DEDUP; } dataref.ref_guid = drrw->drr_toguid; dataref.ref_object = drrw->drr_object; dataref.ref_offset = drrw->drr_offset; if (ddt_update(dda->dedup_hdl, &ddt, &drrw->drr_key.ddk_cksum, drrw->drr_key.ddk_prop, &dataref)) { + dmu_replay_record_t wbr_drr = {0}; + struct drr_write_byref *wbr_drrr = + &wbr_drr.drr_u.drr_write_byref; + /* block already present in stream */ + wbr_drr.drr_type = DRR_WRITE_BYREF; + wbr_drrr->drr_object = drrw->drr_object; wbr_drrr->drr_offset = drrw->drr_offset; wbr_drrr->drr_length = drrw->drr_length; wbr_drrr->drr_toguid = drrw->drr_toguid; wbr_drrr->drr_refguid = dataref.ref_guid; wbr_drrr->drr_refobject = dataref.ref_object; wbr_drrr->drr_refoffset = dataref.ref_offset; wbr_drrr->drr_checksumtype = drrw->drr_checksumtype; wbr_drrr->drr_checksumflags = drrw->drr_checksumtype; wbr_drrr->drr_key.ddk_cksum = drrw->drr_key.ddk_cksum; wbr_drrr->drr_key.ddk_prop = drrw->drr_key.ddk_prop; - if (cksum_and_write(&wbr_drr, - sizeof (dmu_replay_record_t), &stream_cksum, - outfd) == -1) + if (dump_record(&wbr_drr, NULL, 0, + &stream_cksum, outfd) != 0) goto out; } else { /* block not previously seen */ - if (cksum_and_write(drr, - sizeof (dmu_replay_record_t), &stream_cksum, - outfd) == -1) + if (dump_record(drr, buf, drrw->drr_length, + &stream_cksum, outfd) != 0) goto out; - if (cksum_and_write(buf, - drrw->drr_length, - &stream_cksum, outfd) == -1) - goto out; } break; } case DRR_WRITE_EMBEDDED: { - if (cksum_and_write(drr, sizeof (dmu_replay_record_t), - &stream_cksum, outfd) == -1) - goto out; + struct drr_write_embedded *drrwe = + &drr->drr_u.drr_write_embedded; (void) ssread(buf, P2ROUNDUP((uint64_t)drrwe->drr_psize, 8), ofp); - if (cksum_and_write(buf, + if (dump_record(drr, buf, P2ROUNDUP((uint64_t)drrwe->drr_psize, 8), - &stream_cksum, outfd) == -1) + &stream_cksum, outfd) != 0) goto out; break; } case DRR_FREE: { - if (cksum_and_write(drr, sizeof (dmu_replay_record_t), - &stream_cksum, outfd) == -1) + if (dump_record(drr, NULL, 0, &stream_cksum, + outfd) != 0) goto out; break; } default: - (void) printf("INVALID record type 0x%x\n", + (void) fprintf(stderr, "INVALID record type 0x%x\n", drr->drr_type); /* should never happen, so assert */ assert(B_FALSE); } } out: umem_cache_destroy(ddt.ddecache); free(ddt.dedup_hash_array); free(buf); (void) fclose(ofp); return (NULL); } /* * Routines for dealing with the AVL tree of fs-nvlists */ typedef struct fsavl_node { avl_node_t fn_node; nvlist_t *fn_nvfs; char *fn_snapname; uint64_t fn_guid; } fsavl_node_t; static int fsavl_compare(const void *arg1, const void *arg2) { const fsavl_node_t *fn1 = arg1; const fsavl_node_t *fn2 = arg2; if (fn1->fn_guid > fn2->fn_guid) return (+1); else if (fn1->fn_guid < fn2->fn_guid) return (-1); else return (0); } /* * Given the GUID of a snapshot, find its containing filesystem and * (optionally) name. */ static nvlist_t * fsavl_find(avl_tree_t *avl, uint64_t snapguid, char **snapname) { fsavl_node_t fn_find; fsavl_node_t *fn; fn_find.fn_guid = snapguid; fn = avl_find(avl, &fn_find, NULL); if (fn) { if (snapname) *snapname = fn->fn_snapname; return (fn->fn_nvfs); } return (NULL); } static void fsavl_destroy(avl_tree_t *avl) { fsavl_node_t *fn; void *cookie; if (avl == NULL) return; cookie = NULL; while ((fn = avl_destroy_nodes(avl, &cookie)) != NULL) free(fn); avl_destroy(avl); free(avl); } /* * Given an nvlist, produce an avl tree of snapshots, ordered by guid */ static avl_tree_t * fsavl_create(nvlist_t *fss) { avl_tree_t *fsavl; nvpair_t *fselem = NULL; if ((fsavl = malloc(sizeof (avl_tree_t))) == NULL) return (NULL); avl_create(fsavl, fsavl_compare, sizeof (fsavl_node_t), offsetof(fsavl_node_t, fn_node)); while ((fselem = nvlist_next_nvpair(fss, fselem)) != NULL) { nvlist_t *nvfs, *snaps; nvpair_t *snapelem = NULL; VERIFY(0 == nvpair_value_nvlist(fselem, &nvfs)); VERIFY(0 == nvlist_lookup_nvlist(nvfs, "snaps", &snaps)); while ((snapelem = nvlist_next_nvpair(snaps, snapelem)) != NULL) { fsavl_node_t *fn; uint64_t guid; VERIFY(0 == nvpair_value_uint64(snapelem, &guid)); if ((fn = malloc(sizeof (fsavl_node_t))) == NULL) { fsavl_destroy(fsavl); return (NULL); } fn->fn_nvfs = nvfs; fn->fn_snapname = nvpair_name(snapelem); fn->fn_guid = guid; /* * Note: if there are multiple snaps with the * same GUID, we ignore all but one. */ if (avl_find(fsavl, fn, NULL) == NULL) avl_add(fsavl, fn); else free(fn); } } return (fsavl); } /* * Routines for dealing with the giant nvlist of fs-nvlists, etc. */ typedef struct send_data { uint64_t parent_fromsnap_guid; nvlist_t *parent_snaps; nvlist_t *fss; nvlist_t *snapprops; const char *fromsnap; const char *tosnap; boolean_t recursive; /* * The header nvlist is of the following format: * { * "tosnap" -> string * "fromsnap" -> string (if incremental) * "fss" -> { * id -> { * * "name" -> string (full name; for debugging) * "parentfromsnap" -> number (guid of fromsnap in parent) * * "props" -> { name -> value (only if set here) } * "snaps" -> { name (lastname) -> number (guid) } * "snapprops" -> { name (lastname) -> { name -> value } } * * "origin" -> number (guid) (if clone) * "sent" -> boolean (not on-disk) * } * } * } * */ } send_data_t; static void send_iterate_prop(zfs_handle_t *zhp, nvlist_t *nv); static int send_iterate_snap(zfs_handle_t *zhp, void *arg) { send_data_t *sd = arg; uint64_t guid = zhp->zfs_dmustats.dds_guid; char *snapname; nvlist_t *nv; snapname = strrchr(zhp->zfs_name, '@')+1; VERIFY(0 == nvlist_add_uint64(sd->parent_snaps, snapname, guid)); /* * NB: if there is no fromsnap here (it's a newly created fs in * an incremental replication), we will substitute the tosnap. */ if ((sd->fromsnap && strcmp(snapname, sd->fromsnap) == 0) || (sd->parent_fromsnap_guid == 0 && sd->tosnap && strcmp(snapname, sd->tosnap) == 0)) { sd->parent_fromsnap_guid = guid; } VERIFY(0 == nvlist_alloc(&nv, NV_UNIQUE_NAME, 0)); send_iterate_prop(zhp, nv); VERIFY(0 == nvlist_add_nvlist(sd->snapprops, snapname, nv)); nvlist_free(nv); zfs_close(zhp); return (0); } static void send_iterate_prop(zfs_handle_t *zhp, nvlist_t *nv) { nvpair_t *elem = NULL; while ((elem = nvlist_next_nvpair(zhp->zfs_props, elem)) != NULL) { char *propname = nvpair_name(elem); zfs_prop_t prop = zfs_name_to_prop(propname); nvlist_t *propnv; if (!zfs_prop_user(propname)) { /* * Realistically, this should never happen. However, * we want the ability to add DSL properties without * needing to make incompatible version changes. We * need to ignore unknown properties to allow older * software to still send datasets containing these * properties, with the unknown properties elided. */ if (prop == ZPROP_INVAL) continue; if (zfs_prop_readonly(prop)) continue; } verify(nvpair_value_nvlist(elem, &propnv) == 0); if (prop == ZFS_PROP_QUOTA || prop == ZFS_PROP_RESERVATION || prop == ZFS_PROP_REFQUOTA || prop == ZFS_PROP_REFRESERVATION) { char *source; uint64_t value; verify(nvlist_lookup_uint64(propnv, ZPROP_VALUE, &value) == 0); if (zhp->zfs_type == ZFS_TYPE_SNAPSHOT) continue; /* * May have no source before SPA_VERSION_RECVD_PROPS, * but is still modifiable. */ if (nvlist_lookup_string(propnv, ZPROP_SOURCE, &source) == 0) { if ((strcmp(source, zhp->zfs_name) != 0) && (strcmp(source, ZPROP_SOURCE_VAL_RECVD) != 0)) continue; } } else { char *source; if (nvlist_lookup_string(propnv, ZPROP_SOURCE, &source) != 0) continue; if ((strcmp(source, zhp->zfs_name) != 0) && (strcmp(source, ZPROP_SOURCE_VAL_RECVD) != 0)) continue; } if (zfs_prop_user(propname) || zfs_prop_get_type(prop) == PROP_TYPE_STRING) { char *value; verify(nvlist_lookup_string(propnv, ZPROP_VALUE, &value) == 0); VERIFY(0 == nvlist_add_string(nv, propname, value)); } else { uint64_t value; verify(nvlist_lookup_uint64(propnv, ZPROP_VALUE, &value) == 0); VERIFY(0 == nvlist_add_uint64(nv, propname, value)); } } } /* * recursively generate nvlists describing datasets. See comment * for the data structure send_data_t above for description of contents * of the nvlist. */ static int send_iterate_fs(zfs_handle_t *zhp, void *arg) { send_data_t *sd = arg; nvlist_t *nvfs, *nv; int rv = 0; uint64_t parent_fromsnap_guid_save = sd->parent_fromsnap_guid; uint64_t guid = zhp->zfs_dmustats.dds_guid; char guidstring[64]; VERIFY(0 == nvlist_alloc(&nvfs, NV_UNIQUE_NAME, 0)); VERIFY(0 == nvlist_add_string(nvfs, "name", zhp->zfs_name)); VERIFY(0 == nvlist_add_uint64(nvfs, "parentfromsnap", sd->parent_fromsnap_guid)); if (zhp->zfs_dmustats.dds_origin[0]) { zfs_handle_t *origin = zfs_open(zhp->zfs_hdl, zhp->zfs_dmustats.dds_origin, ZFS_TYPE_SNAPSHOT); if (origin == NULL) return (-1); VERIFY(0 == nvlist_add_uint64(nvfs, "origin", origin->zfs_dmustats.dds_guid)); } /* iterate over props */ VERIFY(0 == nvlist_alloc(&nv, NV_UNIQUE_NAME, 0)); send_iterate_prop(zhp, nv); VERIFY(0 == nvlist_add_nvlist(nvfs, "props", nv)); nvlist_free(nv); /* iterate over snaps, and set sd->parent_fromsnap_guid */ sd->parent_fromsnap_guid = 0; VERIFY(0 == nvlist_alloc(&sd->parent_snaps, NV_UNIQUE_NAME, 0)); VERIFY(0 == nvlist_alloc(&sd->snapprops, NV_UNIQUE_NAME, 0)); (void) zfs_iter_snapshots_sorted(zhp, send_iterate_snap, sd); VERIFY(0 == nvlist_add_nvlist(nvfs, "snaps", sd->parent_snaps)); VERIFY(0 == nvlist_add_nvlist(nvfs, "snapprops", sd->snapprops)); nvlist_free(sd->parent_snaps); nvlist_free(sd->snapprops); /* add this fs to nvlist */ (void) snprintf(guidstring, sizeof (guidstring), "0x%llx", (longlong_t)guid); VERIFY(0 == nvlist_add_nvlist(sd->fss, guidstring, nvfs)); nvlist_free(nvfs); /* iterate over children */ if (sd->recursive) rv = zfs_iter_filesystems(zhp, send_iterate_fs, sd); sd->parent_fromsnap_guid = parent_fromsnap_guid_save; zfs_close(zhp); return (rv); } static int gather_nvlist(libzfs_handle_t *hdl, const char *fsname, const char *fromsnap, const char *tosnap, boolean_t recursive, nvlist_t **nvlp, avl_tree_t **avlp) { zfs_handle_t *zhp; send_data_t sd = { 0 }; int error; zhp = zfs_open(hdl, fsname, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME); if (zhp == NULL) return (EZFS_BADTYPE); VERIFY(0 == nvlist_alloc(&sd.fss, NV_UNIQUE_NAME, 0)); sd.fromsnap = fromsnap; sd.tosnap = tosnap; sd.recursive = recursive; if ((error = send_iterate_fs(zhp, &sd)) != 0) { nvlist_free(sd.fss); if (avlp != NULL) *avlp = NULL; *nvlp = NULL; return (error); } if (avlp != NULL && (*avlp = fsavl_create(sd.fss)) == NULL) { nvlist_free(sd.fss); *nvlp = NULL; return (EZFS_NOMEM); } *nvlp = sd.fss; return (0); } /* * Routines specific to "zfs send" */ typedef struct send_dump_data { /* these are all just the short snapname (the part after the @) */ const char *fromsnap; const char *tosnap; char prevsnap[ZFS_MAXNAMELEN]; uint64_t prevsnap_obj; boolean_t seenfrom, seento, replicate, doall, fromorigin; boolean_t verbose, dryrun, parsable, progress, embed_data, large_block; int outfd; boolean_t err; nvlist_t *fss; nvlist_t *snapholds; avl_tree_t *fsavl; snapfilter_cb_t *filter_cb; void *filter_cb_arg; nvlist_t *debugnv; char holdtag[ZFS_MAXNAMELEN]; int cleanup_fd; uint64_t size; } send_dump_data_t; static int estimate_ioctl(zfs_handle_t *zhp, uint64_t fromsnap_obj, boolean_t fromorigin, uint64_t *sizep) { zfs_cmd_t zc = { 0 }; libzfs_handle_t *hdl = zhp->zfs_hdl; assert(zhp->zfs_type == ZFS_TYPE_SNAPSHOT); assert(fromsnap_obj == 0 || !fromorigin); (void) strlcpy(zc.zc_name, zhp->zfs_name, sizeof (zc.zc_name)); zc.zc_obj = fromorigin; zc.zc_sendobj = zfs_prop_get_int(zhp, ZFS_PROP_OBJSETID); zc.zc_fromobj = fromsnap_obj; zc.zc_guid = 1; /* estimate flag */ if (zfs_ioctl(zhp->zfs_hdl, ZFS_IOC_SEND, &zc) != 0) { char errbuf[1024]; (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "warning: cannot estimate space for '%s'"), zhp->zfs_name); switch (errno) { case EXDEV: zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "not an earlier snapshot from the same fs")); return (zfs_error(hdl, EZFS_CROSSTARGET, errbuf)); case ENOENT: if (zfs_dataset_exists(hdl, zc.zc_name, ZFS_TYPE_SNAPSHOT)) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "incremental source (@%s) does not exist"), zc.zc_value); } return (zfs_error(hdl, EZFS_NOENT, errbuf)); case EDQUOT: case EFBIG: case EIO: case ENOLINK: case ENOSPC: case ENXIO: case EPIPE: case ERANGE: case EFAULT: case EROFS: zfs_error_aux(hdl, strerror(errno)); return (zfs_error(hdl, EZFS_BADBACKUP, errbuf)); default: return (zfs_standard_error(hdl, errno, errbuf)); } } *sizep = zc.zc_objset_type; return (0); } /* * Dumps a backup of the given snapshot (incremental from fromsnap if it's not * NULL) to the file descriptor specified by outfd. */ static int dump_ioctl(zfs_handle_t *zhp, const char *fromsnap, uint64_t fromsnap_obj, boolean_t fromorigin, int outfd, enum lzc_send_flags flags, nvlist_t *debugnv) { zfs_cmd_t zc = { 0 }; libzfs_handle_t *hdl = zhp->zfs_hdl; nvlist_t *thisdbg; assert(zhp->zfs_type == ZFS_TYPE_SNAPSHOT); assert(fromsnap_obj == 0 || !fromorigin); (void) strlcpy(zc.zc_name, zhp->zfs_name, sizeof (zc.zc_name)); zc.zc_cookie = outfd; zc.zc_obj = fromorigin; zc.zc_sendobj = zfs_prop_get_int(zhp, ZFS_PROP_OBJSETID); zc.zc_fromobj = fromsnap_obj; zc.zc_flags = flags; VERIFY(0 == nvlist_alloc(&thisdbg, NV_UNIQUE_NAME, 0)); if (fromsnap && fromsnap[0] != '\0') { VERIFY(0 == nvlist_add_string(thisdbg, "fromsnap", fromsnap)); } if (zfs_ioctl(zhp->zfs_hdl, ZFS_IOC_SEND, &zc) != 0) { char errbuf[1024]; (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "warning: cannot send '%s'"), zhp->zfs_name); VERIFY(0 == nvlist_add_uint64(thisdbg, "error", errno)); if (debugnv) { VERIFY(0 == nvlist_add_nvlist(debugnv, zhp->zfs_name, thisdbg)); } nvlist_free(thisdbg); switch (errno) { case EXDEV: zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "not an earlier snapshot from the same fs")); return (zfs_error(hdl, EZFS_CROSSTARGET, errbuf)); case ENOENT: if (zfs_dataset_exists(hdl, zc.zc_name, ZFS_TYPE_SNAPSHOT)) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "incremental source (@%s) does not exist"), zc.zc_value); } return (zfs_error(hdl, EZFS_NOENT, errbuf)); case EDQUOT: case EFBIG: case EIO: case ENOLINK: case ENOSPC: #ifdef illumos case ENOSTR: #endif case ENXIO: case EPIPE: case ERANGE: case EFAULT: case EROFS: zfs_error_aux(hdl, strerror(errno)); return (zfs_error(hdl, EZFS_BADBACKUP, errbuf)); default: return (zfs_standard_error(hdl, errno, errbuf)); } } if (debugnv) VERIFY(0 == nvlist_add_nvlist(debugnv, zhp->zfs_name, thisdbg)); nvlist_free(thisdbg); return (0); } static void gather_holds(zfs_handle_t *zhp, send_dump_data_t *sdd) { assert(zhp->zfs_type == ZFS_TYPE_SNAPSHOT); /* * zfs_send() only sets snapholds for sends that need them, * e.g. replication and doall. */ if (sdd->snapholds == NULL) return; fnvlist_add_string(sdd->snapholds, zhp->zfs_name, sdd->holdtag); } static void * send_progress_thread(void *arg) { progress_arg_t *pa = arg; zfs_cmd_t zc = { 0 }; zfs_handle_t *zhp = pa->pa_zhp; libzfs_handle_t *hdl = zhp->zfs_hdl; unsigned long long bytes; char buf[16]; time_t t; struct tm *tm; assert(zhp->zfs_type == ZFS_TYPE_SNAPSHOT); (void) strlcpy(zc.zc_name, zhp->zfs_name, sizeof (zc.zc_name)); if (!pa->pa_parsable) (void) fprintf(stderr, "TIME SENT SNAPSHOT\n"); /* * Print the progress from ZFS_IOC_SEND_PROGRESS every second. */ for (;;) { (void) sleep(1); zc.zc_cookie = pa->pa_fd; if (zfs_ioctl(hdl, ZFS_IOC_SEND_PROGRESS, &zc) != 0) return ((void *)-1); (void) time(&t); tm = localtime(&t); bytes = zc.zc_cookie; if (pa->pa_parsable) { (void) fprintf(stderr, "%02d:%02d:%02d\t%llu\t%s\n", tm->tm_hour, tm->tm_min, tm->tm_sec, bytes, zhp->zfs_name); } else { zfs_nicenum(bytes, buf, sizeof (buf)); (void) fprintf(stderr, "%02d:%02d:%02d %5s %s\n", tm->tm_hour, tm->tm_min, tm->tm_sec, buf, zhp->zfs_name); } } } static int dump_snapshot(zfs_handle_t *zhp, void *arg) { send_dump_data_t *sdd = arg; progress_arg_t pa = { 0 }; pthread_t tid; char *thissnap; int err; boolean_t isfromsnap, istosnap, fromorigin; boolean_t exclude = B_FALSE; err = 0; thissnap = strchr(zhp->zfs_name, '@') + 1; isfromsnap = (sdd->fromsnap != NULL && strcmp(sdd->fromsnap, thissnap) == 0); if (!sdd->seenfrom && isfromsnap) { gather_holds(zhp, sdd); sdd->seenfrom = B_TRUE; (void) strcpy(sdd->prevsnap, thissnap); sdd->prevsnap_obj = zfs_prop_get_int(zhp, ZFS_PROP_OBJSETID); zfs_close(zhp); return (0); } if (sdd->seento || !sdd->seenfrom) { zfs_close(zhp); return (0); } istosnap = (strcmp(sdd->tosnap, thissnap) == 0); if (istosnap) sdd->seento = B_TRUE; if (!sdd->doall && !isfromsnap && !istosnap) { if (sdd->replicate) { char *snapname; nvlist_t *snapprops; /* * Filter out all intermediate snapshots except origin * snapshots needed to replicate clones. */ nvlist_t *nvfs = fsavl_find(sdd->fsavl, zhp->zfs_dmustats.dds_guid, &snapname); VERIFY(0 == nvlist_lookup_nvlist(nvfs, "snapprops", &snapprops)); VERIFY(0 == nvlist_lookup_nvlist(snapprops, thissnap, &snapprops)); exclude = !nvlist_exists(snapprops, "is_clone_origin"); } else { exclude = B_TRUE; } } /* * If a filter function exists, call it to determine whether * this snapshot will be sent. */ if (exclude || (sdd->filter_cb != NULL && sdd->filter_cb(zhp, sdd->filter_cb_arg) == B_FALSE)) { /* * This snapshot is filtered out. Don't send it, and don't * set prevsnap_obj, so it will be as if this snapshot didn't * exist, and the next accepted snapshot will be sent as * an incremental from the last accepted one, or as the * first (and full) snapshot in the case of a replication, * non-incremental send. */ zfs_close(zhp); return (0); } gather_holds(zhp, sdd); fromorigin = sdd->prevsnap[0] == '\0' && (sdd->fromorigin || sdd->replicate); if (sdd->verbose) { uint64_t size; err = estimate_ioctl(zhp, sdd->prevsnap_obj, fromorigin, &size); if (sdd->parsable) { if (sdd->prevsnap[0] != '\0') { (void) fprintf(stderr, "incremental\t%s\t%s", sdd->prevsnap, zhp->zfs_name); } else { (void) fprintf(stderr, "full\t%s", zhp->zfs_name); } } else { (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "send from @%s to %s"), sdd->prevsnap, zhp->zfs_name); } if (err == 0) { if (sdd->parsable) { (void) fprintf(stderr, "\t%llu\n", (longlong_t)size); } else { char buf[16]; zfs_nicenum(size, buf, sizeof (buf)); (void) fprintf(stderr, dgettext(TEXT_DOMAIN, " estimated size is %s\n"), buf); } sdd->size += size; } else { (void) fprintf(stderr, "\n"); } } if (!sdd->dryrun) { /* * If progress reporting is requested, spawn a new thread to * poll ZFS_IOC_SEND_PROGRESS at a regular interval. */ if (sdd->progress) { pa.pa_zhp = zhp; pa.pa_fd = sdd->outfd; pa.pa_parsable = sdd->parsable; if (err = pthread_create(&tid, NULL, send_progress_thread, &pa)) { zfs_close(zhp); return (err); } } enum lzc_send_flags flags = 0; if (sdd->large_block) flags |= LZC_SEND_FLAG_LARGE_BLOCK; if (sdd->embed_data) flags |= LZC_SEND_FLAG_EMBED_DATA; err = dump_ioctl(zhp, sdd->prevsnap, sdd->prevsnap_obj, fromorigin, sdd->outfd, flags, sdd->debugnv); if (sdd->progress) { (void) pthread_cancel(tid); (void) pthread_join(tid, NULL); } } (void) strcpy(sdd->prevsnap, thissnap); sdd->prevsnap_obj = zfs_prop_get_int(zhp, ZFS_PROP_OBJSETID); zfs_close(zhp); return (err); } static int dump_filesystem(zfs_handle_t *zhp, void *arg) { int rv = 0; send_dump_data_t *sdd = arg; boolean_t missingfrom = B_FALSE; zfs_cmd_t zc = { 0 }; (void) snprintf(zc.zc_name, sizeof (zc.zc_name), "%s@%s", zhp->zfs_name, sdd->tosnap); if (ioctl(zhp->zfs_hdl->libzfs_fd, ZFS_IOC_OBJSET_STATS, &zc) != 0) { (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "WARNING: could not send %s@%s: does not exist\n"), zhp->zfs_name, sdd->tosnap); sdd->err = B_TRUE; return (0); } if (sdd->replicate && sdd->fromsnap) { /* * If this fs does not have fromsnap, and we're doing * recursive, we need to send a full stream from the * beginning (or an incremental from the origin if this * is a clone). If we're doing non-recursive, then let * them get the error. */ (void) snprintf(zc.zc_name, sizeof (zc.zc_name), "%s@%s", zhp->zfs_name, sdd->fromsnap); if (ioctl(zhp->zfs_hdl->libzfs_fd, ZFS_IOC_OBJSET_STATS, &zc) != 0) { missingfrom = B_TRUE; } } sdd->seenfrom = sdd->seento = sdd->prevsnap[0] = 0; sdd->prevsnap_obj = 0; if (sdd->fromsnap == NULL || missingfrom) sdd->seenfrom = B_TRUE; rv = zfs_iter_snapshots_sorted(zhp, dump_snapshot, arg); if (!sdd->seenfrom) { (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "WARNING: could not send %s@%s:\n" "incremental source (%s@%s) does not exist\n"), zhp->zfs_name, sdd->tosnap, zhp->zfs_name, sdd->fromsnap); sdd->err = B_TRUE; } else if (!sdd->seento) { if (sdd->fromsnap) { (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "WARNING: could not send %s@%s:\n" "incremental source (%s@%s) " "is not earlier than it\n"), zhp->zfs_name, sdd->tosnap, zhp->zfs_name, sdd->fromsnap); } else { (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "WARNING: " "could not send %s@%s: does not exist\n"), zhp->zfs_name, sdd->tosnap); } sdd->err = B_TRUE; } return (rv); } static int dump_filesystems(zfs_handle_t *rzhp, void *arg) { send_dump_data_t *sdd = arg; nvpair_t *fspair; boolean_t needagain, progress; if (!sdd->replicate) return (dump_filesystem(rzhp, sdd)); /* Mark the clone origin snapshots. */ for (fspair = nvlist_next_nvpair(sdd->fss, NULL); fspair; fspair = nvlist_next_nvpair(sdd->fss, fspair)) { nvlist_t *nvfs; uint64_t origin_guid = 0; VERIFY(0 == nvpair_value_nvlist(fspair, &nvfs)); (void) nvlist_lookup_uint64(nvfs, "origin", &origin_guid); if (origin_guid != 0) { char *snapname; nvlist_t *origin_nv = fsavl_find(sdd->fsavl, origin_guid, &snapname); if (origin_nv != NULL) { nvlist_t *snapprops; VERIFY(0 == nvlist_lookup_nvlist(origin_nv, "snapprops", &snapprops)); VERIFY(0 == nvlist_lookup_nvlist(snapprops, snapname, &snapprops)); VERIFY(0 == nvlist_add_boolean( snapprops, "is_clone_origin")); } } } again: needagain = progress = B_FALSE; for (fspair = nvlist_next_nvpair(sdd->fss, NULL); fspair; fspair = nvlist_next_nvpair(sdd->fss, fspair)) { nvlist_t *fslist, *parent_nv; char *fsname; zfs_handle_t *zhp; int err; uint64_t origin_guid = 0; uint64_t parent_guid = 0; VERIFY(nvpair_value_nvlist(fspair, &fslist) == 0); if (nvlist_lookup_boolean(fslist, "sent") == 0) continue; VERIFY(nvlist_lookup_string(fslist, "name", &fsname) == 0); (void) nvlist_lookup_uint64(fslist, "origin", &origin_guid); (void) nvlist_lookup_uint64(fslist, "parentfromsnap", &parent_guid); if (parent_guid != 0) { parent_nv = fsavl_find(sdd->fsavl, parent_guid, NULL); if (!nvlist_exists(parent_nv, "sent")) { /* parent has not been sent; skip this one */ needagain = B_TRUE; continue; } } if (origin_guid != 0) { nvlist_t *origin_nv = fsavl_find(sdd->fsavl, origin_guid, NULL); if (origin_nv != NULL && !nvlist_exists(origin_nv, "sent")) { /* * origin has not been sent yet; * skip this clone. */ needagain = B_TRUE; continue; } } zhp = zfs_open(rzhp->zfs_hdl, fsname, ZFS_TYPE_DATASET); if (zhp == NULL) return (-1); err = dump_filesystem(zhp, sdd); VERIFY(nvlist_add_boolean(fslist, "sent") == 0); progress = B_TRUE; zfs_close(zhp); if (err) return (err); } if (needagain) { assert(progress); goto again; } /* clean out the sent flags in case we reuse this fss */ for (fspair = nvlist_next_nvpair(sdd->fss, NULL); fspair; fspair = nvlist_next_nvpair(sdd->fss, fspair)) { nvlist_t *fslist; VERIFY(nvpair_value_nvlist(fspair, &fslist) == 0); (void) nvlist_remove_all(fslist, "sent"); } return (0); } /* * Generate a send stream for the dataset identified by the argument zhp. * * The content of the send stream is the snapshot identified by * 'tosnap'. Incremental streams are requested in two ways: * - from the snapshot identified by "fromsnap" (if non-null) or * - from the origin of the dataset identified by zhp, which must * be a clone. In this case, "fromsnap" is null and "fromorigin" * is TRUE. * * The send stream is recursive (i.e. dumps a hierarchy of snapshots) and * uses a special header (with a hdrtype field of DMU_COMPOUNDSTREAM) * if "replicate" is set. If "doall" is set, dump all the intermediate * snapshots. The DMU_COMPOUNDSTREAM header is used in the "doall" * case too. If "props" is set, send properties. */ int zfs_send(zfs_handle_t *zhp, const char *fromsnap, const char *tosnap, sendflags_t *flags, int outfd, snapfilter_cb_t filter_func, void *cb_arg, nvlist_t **debugnvp) { char errbuf[1024]; send_dump_data_t sdd = { 0 }; int err = 0; nvlist_t *fss = NULL; avl_tree_t *fsavl = NULL; static uint64_t holdseq; int spa_version; pthread_t tid = 0; int pipefd[2]; dedup_arg_t dda = { 0 }; int featureflags = 0; (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "cannot send '%s'"), zhp->zfs_name); if (fromsnap && fromsnap[0] == '\0') { zfs_error_aux(zhp->zfs_hdl, dgettext(TEXT_DOMAIN, "zero-length incremental source")); return (zfs_error(zhp->zfs_hdl, EZFS_NOENT, errbuf)); } if (zhp->zfs_type == ZFS_TYPE_FILESYSTEM) { uint64_t version; version = zfs_prop_get_int(zhp, ZFS_PROP_VERSION); if (version >= ZPL_VERSION_SA) { featureflags |= DMU_BACKUP_FEATURE_SA_SPILL; } } if (flags->dedup && !flags->dryrun) { featureflags |= (DMU_BACKUP_FEATURE_DEDUP | DMU_BACKUP_FEATURE_DEDUPPROPS); if (err = pipe(pipefd)) { zfs_error_aux(zhp->zfs_hdl, strerror(errno)); return (zfs_error(zhp->zfs_hdl, EZFS_PIPEFAILED, errbuf)); } dda.outputfd = outfd; dda.inputfd = pipefd[1]; dda.dedup_hdl = zhp->zfs_hdl; if (err = pthread_create(&tid, NULL, cksummer, &dda)) { (void) close(pipefd[0]); (void) close(pipefd[1]); zfs_error_aux(zhp->zfs_hdl, strerror(errno)); return (zfs_error(zhp->zfs_hdl, EZFS_THREADCREATEFAILED, errbuf)); } } if (flags->replicate || flags->doall || flags->props) { dmu_replay_record_t drr = { 0 }; char *packbuf = NULL; size_t buflen = 0; zio_cksum_t zc = { 0 }; if (flags->replicate || flags->props) { nvlist_t *hdrnv; VERIFY(0 == nvlist_alloc(&hdrnv, NV_UNIQUE_NAME, 0)); if (fromsnap) { VERIFY(0 == nvlist_add_string(hdrnv, "fromsnap", fromsnap)); } VERIFY(0 == nvlist_add_string(hdrnv, "tosnap", tosnap)); if (!flags->replicate) { VERIFY(0 == nvlist_add_boolean(hdrnv, "not_recursive")); } err = gather_nvlist(zhp->zfs_hdl, zhp->zfs_name, fromsnap, tosnap, flags->replicate, &fss, &fsavl); if (err) goto err_out; VERIFY(0 == nvlist_add_nvlist(hdrnv, "fss", fss)); err = nvlist_pack(hdrnv, &packbuf, &buflen, NV_ENCODE_XDR, 0); if (debugnvp) *debugnvp = hdrnv; else nvlist_free(hdrnv); if (err) goto stderr_out; } if (!flags->dryrun) { /* write first begin record */ drr.drr_type = DRR_BEGIN; drr.drr_u.drr_begin.drr_magic = DMU_BACKUP_MAGIC; DMU_SET_STREAM_HDRTYPE(drr.drr_u.drr_begin. drr_versioninfo, DMU_COMPOUNDSTREAM); DMU_SET_FEATUREFLAGS(drr.drr_u.drr_begin. drr_versioninfo, featureflags); (void) snprintf(drr.drr_u.drr_begin.drr_toname, sizeof (drr.drr_u.drr_begin.drr_toname), "%s@%s", zhp->zfs_name, tosnap); drr.drr_payloadlen = buflen; - err = cksum_and_write(&drr, sizeof (drr), &zc, outfd); - /* write header nvlist */ - if (err != -1 && packbuf != NULL) { - err = cksum_and_write(packbuf, buflen, &zc, - outfd); - } + err = dump_record(&drr, packbuf, buflen, &zc, outfd); free(packbuf); - if (err == -1) { - err = errno; + if (err != 0) goto stderr_out; - } /* write end record */ bzero(&drr, sizeof (drr)); drr.drr_type = DRR_END; drr.drr_u.drr_end.drr_checksum = zc; err = write(outfd, &drr, sizeof (drr)); if (err == -1) { err = errno; goto stderr_out; } err = 0; } } /* dump each stream */ sdd.fromsnap = fromsnap; sdd.tosnap = tosnap; if (tid != 0) sdd.outfd = pipefd[0]; else sdd.outfd = outfd; sdd.replicate = flags->replicate; sdd.doall = flags->doall; sdd.fromorigin = flags->fromorigin; sdd.fss = fss; sdd.fsavl = fsavl; sdd.verbose = flags->verbose; sdd.parsable = flags->parsable; sdd.progress = flags->progress; sdd.dryrun = flags->dryrun; sdd.large_block = flags->largeblock; sdd.embed_data = flags->embed_data; sdd.filter_cb = filter_func; sdd.filter_cb_arg = cb_arg; if (debugnvp) sdd.debugnv = *debugnvp; /* * Some flags require that we place user holds on the datasets that are * being sent so they don't get destroyed during the send. We can skip * this step if the pool is imported read-only since the datasets cannot * be destroyed. */ if (!flags->dryrun && !zpool_get_prop_int(zfs_get_pool_handle(zhp), ZPOOL_PROP_READONLY, NULL) && zfs_spa_version(zhp, &spa_version) == 0 && spa_version >= SPA_VERSION_USERREFS && (flags->doall || flags->replicate)) { ++holdseq; (void) snprintf(sdd.holdtag, sizeof (sdd.holdtag), ".send-%d-%llu", getpid(), (u_longlong_t)holdseq); sdd.cleanup_fd = open(ZFS_DEV, O_RDWR|O_EXCL); if (sdd.cleanup_fd < 0) { err = errno; goto stderr_out; } sdd.snapholds = fnvlist_alloc(); } else { sdd.cleanup_fd = -1; sdd.snapholds = NULL; } if (flags->verbose || sdd.snapholds != NULL) { /* * Do a verbose no-op dry run to get all the verbose output * or to gather snapshot hold's before generating any data, * then do a non-verbose real run to generate the streams. */ sdd.dryrun = B_TRUE; err = dump_filesystems(zhp, &sdd); if (err != 0) goto stderr_out; if (flags->verbose) { if (flags->parsable) { (void) fprintf(stderr, "size\t%llu\n", (longlong_t)sdd.size); } else { char buf[16]; zfs_nicenum(sdd.size, buf, sizeof (buf)); (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "total estimated size is %s\n"), buf); } } /* Ensure no snaps found is treated as an error. */ if (!sdd.seento) { err = ENOENT; goto err_out; } /* Skip the second run if dryrun was requested. */ if (flags->dryrun) goto err_out; if (sdd.snapholds != NULL) { err = zfs_hold_nvl(zhp, sdd.cleanup_fd, sdd.snapholds); if (err != 0) goto stderr_out; fnvlist_free(sdd.snapholds); sdd.snapholds = NULL; } sdd.dryrun = B_FALSE; sdd.verbose = B_FALSE; } err = dump_filesystems(zhp, &sdd); fsavl_destroy(fsavl); nvlist_free(fss); /* Ensure no snaps found is treated as an error. */ if (err == 0 && !sdd.seento) err = ENOENT; if (tid != 0) { if (err != 0) (void) pthread_cancel(tid); (void) close(pipefd[0]); (void) pthread_join(tid, NULL); } if (sdd.cleanup_fd != -1) { VERIFY(0 == close(sdd.cleanup_fd)); sdd.cleanup_fd = -1; } if (!flags->dryrun && (flags->replicate || flags->doall || flags->props)) { /* * write final end record. NB: want to do this even if * there was some error, because it might not be totally * failed. */ dmu_replay_record_t drr = { 0 }; drr.drr_type = DRR_END; if (write(outfd, &drr, sizeof (drr)) == -1) { return (zfs_standard_error(zhp->zfs_hdl, errno, errbuf)); } } return (err || sdd.err); stderr_out: err = zfs_standard_error(zhp->zfs_hdl, err, errbuf); err_out: fsavl_destroy(fsavl); nvlist_free(fss); fnvlist_free(sdd.snapholds); if (sdd.cleanup_fd != -1) VERIFY(0 == close(sdd.cleanup_fd)); if (tid != 0) { (void) pthread_cancel(tid); (void) close(pipefd[0]); (void) pthread_join(tid, NULL); } return (err); } int zfs_send_one(zfs_handle_t *zhp, const char *from, int fd, enum lzc_send_flags flags) { int err; libzfs_handle_t *hdl = zhp->zfs_hdl; char errbuf[1024]; (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "warning: cannot send '%s'"), zhp->zfs_name); err = lzc_send(zhp->zfs_name, from, fd, flags); if (err != 0) { switch (errno) { case EXDEV: zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "not an earlier snapshot from the same fs")); return (zfs_error(hdl, EZFS_CROSSTARGET, errbuf)); case ENOENT: case ESRCH: if (lzc_exists(zhp->zfs_name)) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "incremental source (%s) does not exist"), from); } return (zfs_error(hdl, EZFS_NOENT, errbuf)); case EBUSY: zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "target is busy; if a filesystem, " "it must not be mounted")); return (zfs_error(hdl, EZFS_BUSY, errbuf)); case EDQUOT: case EFBIG: case EIO: case ENOLINK: case ENOSPC: #ifdef illumos case ENOSTR: #endif case ENXIO: case EPIPE: case ERANGE: case EFAULT: case EROFS: zfs_error_aux(hdl, strerror(errno)); return (zfs_error(hdl, EZFS_BADBACKUP, errbuf)); default: return (zfs_standard_error(hdl, errno, errbuf)); } } return (err != 0); } /* * Routines specific to "zfs recv" */ static int recv_read(libzfs_handle_t *hdl, int fd, void *buf, int ilen, boolean_t byteswap, zio_cksum_t *zc) { char *cp = buf; int rv; int len = ilen; + + assert(ilen <= SPA_MAXBLOCKSIZE); do { rv = read(fd, cp, len); cp += rv; len -= rv; } while (rv > 0); if (rv < 0 || len != 0) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "failed to read from stream")); return (zfs_error(hdl, EZFS_BADSTREAM, dgettext(TEXT_DOMAIN, "cannot receive"))); } if (zc) { if (byteswap) fletcher_4_incremental_byteswap(buf, ilen, zc); else fletcher_4_incremental_native(buf, ilen, zc); } return (0); } static int recv_read_nvlist(libzfs_handle_t *hdl, int fd, int len, nvlist_t **nvp, boolean_t byteswap, zio_cksum_t *zc) { char *buf; int err; buf = zfs_alloc(hdl, len); if (buf == NULL) return (ENOMEM); err = recv_read(hdl, fd, buf, len, byteswap, zc); if (err != 0) { free(buf); return (err); } err = nvlist_unpack(buf, len, nvp, 0); free(buf); if (err != 0) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "invalid " "stream (malformed nvlist)")); return (EINVAL); } return (0); } static int recv_rename(libzfs_handle_t *hdl, const char *name, const char *tryname, int baselen, char *newname, recvflags_t *flags) { static int seq; zfs_cmd_t zc = { 0 }; int err; prop_changelist_t *clp; zfs_handle_t *zhp; zhp = zfs_open(hdl, name, ZFS_TYPE_DATASET); if (zhp == NULL) return (-1); clp = changelist_gather(zhp, ZFS_PROP_NAME, 0, flags->force ? MS_FORCE : 0); zfs_close(zhp); if (clp == NULL) return (-1); err = changelist_prefix(clp); if (err) return (err); zc.zc_objset_type = DMU_OST_ZFS; (void) strlcpy(zc.zc_name, name, sizeof (zc.zc_name)); if (tryname) { (void) strcpy(newname, tryname); (void) strlcpy(zc.zc_value, tryname, sizeof (zc.zc_value)); if (flags->verbose) { (void) printf("attempting rename %s to %s\n", zc.zc_name, zc.zc_value); } err = ioctl(hdl->libzfs_fd, ZFS_IOC_RENAME, &zc); if (err == 0) changelist_rename(clp, name, tryname); } else { err = ENOENT; } if (err != 0 && strncmp(name + baselen, "recv-", 5) != 0) { seq++; (void) snprintf(newname, ZFS_MAXNAMELEN, "%.*srecv-%u-%u", baselen, name, getpid(), seq); (void) strlcpy(zc.zc_value, newname, sizeof (zc.zc_value)); if (flags->verbose) { (void) printf("failed - trying rename %s to %s\n", zc.zc_name, zc.zc_value); } err = ioctl(hdl->libzfs_fd, ZFS_IOC_RENAME, &zc); if (err == 0) changelist_rename(clp, name, newname); if (err && flags->verbose) { (void) printf("failed (%u) - " "will try again on next pass\n", errno); } err = EAGAIN; } else if (flags->verbose) { if (err == 0) (void) printf("success\n"); else (void) printf("failed (%u)\n", errno); } (void) changelist_postfix(clp); changelist_free(clp); return (err); } static int recv_destroy(libzfs_handle_t *hdl, const char *name, int baselen, char *newname, recvflags_t *flags) { zfs_cmd_t zc = { 0 }; int err = 0; prop_changelist_t *clp; zfs_handle_t *zhp; boolean_t defer = B_FALSE; int spa_version; zhp = zfs_open(hdl, name, ZFS_TYPE_DATASET); if (zhp == NULL) return (-1); clp = changelist_gather(zhp, ZFS_PROP_NAME, 0, flags->force ? MS_FORCE : 0); if (zfs_get_type(zhp) == ZFS_TYPE_SNAPSHOT && zfs_spa_version(zhp, &spa_version) == 0 && spa_version >= SPA_VERSION_USERREFS) defer = B_TRUE; zfs_close(zhp); if (clp == NULL) return (-1); err = changelist_prefix(clp); if (err) return (err); zc.zc_objset_type = DMU_OST_ZFS; zc.zc_defer_destroy = defer; (void) strlcpy(zc.zc_name, name, sizeof (zc.zc_name)); if (flags->verbose) (void) printf("attempting destroy %s\n", zc.zc_name); err = ioctl(hdl->libzfs_fd, ZFS_IOC_DESTROY, &zc); if (err == 0) { if (flags->verbose) (void) printf("success\n"); changelist_remove(clp, zc.zc_name); } (void) changelist_postfix(clp); changelist_free(clp); /* * Deferred destroy might destroy the snapshot or only mark it to be * destroyed later, and it returns success in either case. */ if (err != 0 || (defer && zfs_dataset_exists(hdl, name, ZFS_TYPE_SNAPSHOT))) { err = recv_rename(hdl, name, NULL, baselen, newname, flags); } return (err); } typedef struct guid_to_name_data { uint64_t guid; char *name; char *skip; } guid_to_name_data_t; static int guid_to_name_cb(zfs_handle_t *zhp, void *arg) { guid_to_name_data_t *gtnd = arg; int err; if (gtnd->skip != NULL && strcmp(zhp->zfs_name, gtnd->skip) == 0) { return (0); } if (zhp->zfs_dmustats.dds_guid == gtnd->guid) { (void) strcpy(gtnd->name, zhp->zfs_name); zfs_close(zhp); return (EEXIST); } err = zfs_iter_children(zhp, guid_to_name_cb, gtnd); zfs_close(zhp); return (err); } /* * Attempt to find the local dataset associated with this guid. In the case of * multiple matches, we attempt to find the "best" match by searching * progressively larger portions of the hierarchy. This allows one to send a * tree of datasets individually and guarantee that we will find the source * guid within that hierarchy, even if there are multiple matches elsewhere. */ static int guid_to_name(libzfs_handle_t *hdl, const char *parent, uint64_t guid, char *name) { /* exhaustive search all local snapshots */ char pname[ZFS_MAXNAMELEN]; guid_to_name_data_t gtnd; int err = 0; zfs_handle_t *zhp; char *cp; gtnd.guid = guid; gtnd.name = name; gtnd.skip = NULL; (void) strlcpy(pname, parent, sizeof (pname)); /* * Search progressively larger portions of the hierarchy. This will * select the "most local" version of the origin snapshot in the case * that there are multiple matching snapshots in the system. */ while ((cp = strrchr(pname, '/')) != NULL) { /* Chop off the last component and open the parent */ *cp = '\0'; zhp = make_dataset_handle(hdl, pname); if (zhp == NULL) continue; err = zfs_iter_children(zhp, guid_to_name_cb, >nd); zfs_close(zhp); if (err == EEXIST) return (0); /* * Remember the dataset that we already searched, so we * skip it next time through. */ gtnd.skip = pname; } return (ENOENT); } /* * Return +1 if guid1 is before guid2, 0 if they are the same, and -1 if * guid1 is after guid2. */ static int created_before(libzfs_handle_t *hdl, avl_tree_t *avl, uint64_t guid1, uint64_t guid2) { nvlist_t *nvfs; char *fsname, *snapname; char buf[ZFS_MAXNAMELEN]; int rv; zfs_handle_t *guid1hdl, *guid2hdl; uint64_t create1, create2; if (guid2 == 0) return (0); if (guid1 == 0) return (1); nvfs = fsavl_find(avl, guid1, &snapname); VERIFY(0 == nvlist_lookup_string(nvfs, "name", &fsname)); (void) snprintf(buf, sizeof (buf), "%s@%s", fsname, snapname); guid1hdl = zfs_open(hdl, buf, ZFS_TYPE_SNAPSHOT); if (guid1hdl == NULL) return (-1); nvfs = fsavl_find(avl, guid2, &snapname); VERIFY(0 == nvlist_lookup_string(nvfs, "name", &fsname)); (void) snprintf(buf, sizeof (buf), "%s@%s", fsname, snapname); guid2hdl = zfs_open(hdl, buf, ZFS_TYPE_SNAPSHOT); if (guid2hdl == NULL) { zfs_close(guid1hdl); return (-1); } create1 = zfs_prop_get_int(guid1hdl, ZFS_PROP_CREATETXG); create2 = zfs_prop_get_int(guid2hdl, ZFS_PROP_CREATETXG); if (create1 < create2) rv = -1; else if (create1 > create2) rv = +1; else rv = 0; zfs_close(guid1hdl); zfs_close(guid2hdl); return (rv); } static int recv_incremental_replication(libzfs_handle_t *hdl, const char *tofs, recvflags_t *flags, nvlist_t *stream_nv, avl_tree_t *stream_avl, nvlist_t *renamed) { nvlist_t *local_nv, *deleted = NULL; avl_tree_t *local_avl; nvpair_t *fselem, *nextfselem; char *fromsnap; char newname[ZFS_MAXNAMELEN]; char guidname[32]; int error; boolean_t needagain, progress, recursive; char *s1, *s2; VERIFY(0 == nvlist_lookup_string(stream_nv, "fromsnap", &fromsnap)); recursive = (nvlist_lookup_boolean(stream_nv, "not_recursive") == ENOENT); if (flags->dryrun) return (0); again: needagain = progress = B_FALSE; VERIFY(0 == nvlist_alloc(&deleted, NV_UNIQUE_NAME, 0)); if ((error = gather_nvlist(hdl, tofs, fromsnap, NULL, recursive, &local_nv, &local_avl)) != 0) return (error); /* * Process deletes and renames */ for (fselem = nvlist_next_nvpair(local_nv, NULL); fselem; fselem = nextfselem) { nvlist_t *nvfs, *snaps; nvlist_t *stream_nvfs = NULL; nvpair_t *snapelem, *nextsnapelem; uint64_t fromguid = 0; uint64_t originguid = 0; uint64_t stream_originguid = 0; uint64_t parent_fromsnap_guid, stream_parent_fromsnap_guid; char *fsname, *stream_fsname; nextfselem = nvlist_next_nvpair(local_nv, fselem); VERIFY(0 == nvpair_value_nvlist(fselem, &nvfs)); VERIFY(0 == nvlist_lookup_nvlist(nvfs, "snaps", &snaps)); VERIFY(0 == nvlist_lookup_string(nvfs, "name", &fsname)); VERIFY(0 == nvlist_lookup_uint64(nvfs, "parentfromsnap", &parent_fromsnap_guid)); (void) nvlist_lookup_uint64(nvfs, "origin", &originguid); /* * First find the stream's fs, so we can check for * a different origin (due to "zfs promote") */ for (snapelem = nvlist_next_nvpair(snaps, NULL); snapelem; snapelem = nvlist_next_nvpair(snaps, snapelem)) { uint64_t thisguid; VERIFY(0 == nvpair_value_uint64(snapelem, &thisguid)); stream_nvfs = fsavl_find(stream_avl, thisguid, NULL); if (stream_nvfs != NULL) break; } /* check for promote */ (void) nvlist_lookup_uint64(stream_nvfs, "origin", &stream_originguid); if (stream_nvfs && originguid != stream_originguid) { switch (created_before(hdl, local_avl, stream_originguid, originguid)) { case 1: { /* promote it! */ zfs_cmd_t zc = { 0 }; nvlist_t *origin_nvfs; char *origin_fsname; if (flags->verbose) (void) printf("promoting %s\n", fsname); origin_nvfs = fsavl_find(local_avl, originguid, NULL); VERIFY(0 == nvlist_lookup_string(origin_nvfs, "name", &origin_fsname)); (void) strlcpy(zc.zc_value, origin_fsname, sizeof (zc.zc_value)); (void) strlcpy(zc.zc_name, fsname, sizeof (zc.zc_name)); error = zfs_ioctl(hdl, ZFS_IOC_PROMOTE, &zc); if (error == 0) progress = B_TRUE; break; } default: break; case -1: fsavl_destroy(local_avl); nvlist_free(local_nv); return (-1); } /* * We had/have the wrong origin, therefore our * list of snapshots is wrong. Need to handle * them on the next pass. */ needagain = B_TRUE; continue; } for (snapelem = nvlist_next_nvpair(snaps, NULL); snapelem; snapelem = nextsnapelem) { uint64_t thisguid; char *stream_snapname; nvlist_t *found, *props; nextsnapelem = nvlist_next_nvpair(snaps, snapelem); VERIFY(0 == nvpair_value_uint64(snapelem, &thisguid)); found = fsavl_find(stream_avl, thisguid, &stream_snapname); /* check for delete */ if (found == NULL) { char name[ZFS_MAXNAMELEN]; if (!flags->force) continue; (void) snprintf(name, sizeof (name), "%s@%s", fsname, nvpair_name(snapelem)); error = recv_destroy(hdl, name, strlen(fsname)+1, newname, flags); if (error) needagain = B_TRUE; else progress = B_TRUE; sprintf(guidname, "%lu", thisguid); nvlist_add_boolean(deleted, guidname); continue; } stream_nvfs = found; if (0 == nvlist_lookup_nvlist(stream_nvfs, "snapprops", &props) && 0 == nvlist_lookup_nvlist(props, stream_snapname, &props)) { zfs_cmd_t zc = { 0 }; zc.zc_cookie = B_TRUE; /* received */ (void) snprintf(zc.zc_name, sizeof (zc.zc_name), "%s@%s", fsname, nvpair_name(snapelem)); if (zcmd_write_src_nvlist(hdl, &zc, props) == 0) { (void) zfs_ioctl(hdl, ZFS_IOC_SET_PROP, &zc); zcmd_free_nvlists(&zc); } } /* check for different snapname */ if (strcmp(nvpair_name(snapelem), stream_snapname) != 0) { char name[ZFS_MAXNAMELEN]; char tryname[ZFS_MAXNAMELEN]; (void) snprintf(name, sizeof (name), "%s@%s", fsname, nvpair_name(snapelem)); (void) snprintf(tryname, sizeof (name), "%s@%s", fsname, stream_snapname); error = recv_rename(hdl, name, tryname, strlen(fsname)+1, newname, flags); if (error) needagain = B_TRUE; else progress = B_TRUE; } if (strcmp(stream_snapname, fromsnap) == 0) fromguid = thisguid; } /* check for delete */ if (stream_nvfs == NULL) { if (!flags->force) continue; error = recv_destroy(hdl, fsname, strlen(tofs)+1, newname, flags); if (error) needagain = B_TRUE; else progress = B_TRUE; sprintf(guidname, "%lu", parent_fromsnap_guid); nvlist_add_boolean(deleted, guidname); continue; } if (fromguid == 0) { if (flags->verbose) { (void) printf("local fs %s does not have " "fromsnap (%s in stream); must have " "been deleted locally; ignoring\n", fsname, fromsnap); } continue; } VERIFY(0 == nvlist_lookup_string(stream_nvfs, "name", &stream_fsname)); VERIFY(0 == nvlist_lookup_uint64(stream_nvfs, "parentfromsnap", &stream_parent_fromsnap_guid)); s1 = strrchr(fsname, '/'); s2 = strrchr(stream_fsname, '/'); /* * Check if we're going to rename based on parent guid change * and the current parent guid was also deleted. If it was then * rename will fail and is likely unneeded, so avoid this and * force an early retry to determine the new * parent_fromsnap_guid. */ if (stream_parent_fromsnap_guid != 0 && parent_fromsnap_guid != 0 && stream_parent_fromsnap_guid != parent_fromsnap_guid) { sprintf(guidname, "%lu", parent_fromsnap_guid); if (nvlist_exists(deleted, guidname)) { progress = B_TRUE; needagain = B_TRUE; goto doagain; } } /* * Check for rename. If the exact receive path is specified, it * does not count as a rename, but we still need to check the * datasets beneath it. */ if ((stream_parent_fromsnap_guid != 0 && parent_fromsnap_guid != 0 && stream_parent_fromsnap_guid != parent_fromsnap_guid) || ((flags->isprefix || strcmp(tofs, fsname) != 0) && (s1 != NULL) && (s2 != NULL) && strcmp(s1, s2) != 0)) { nvlist_t *parent; char tryname[ZFS_MAXNAMELEN]; parent = fsavl_find(local_avl, stream_parent_fromsnap_guid, NULL); /* * NB: parent might not be found if we used the * tosnap for stream_parent_fromsnap_guid, * because the parent is a newly-created fs; * we'll be able to rename it after we recv the * new fs. */ if (parent != NULL) { char *pname; VERIFY(0 == nvlist_lookup_string(parent, "name", &pname)); (void) snprintf(tryname, sizeof (tryname), "%s%s", pname, strrchr(stream_fsname, '/')); } else { tryname[0] = '\0'; if (flags->verbose) { (void) printf("local fs %s new parent " "not found\n", fsname); } } newname[0] = '\0'; error = recv_rename(hdl, fsname, tryname, strlen(tofs)+1, newname, flags); if (renamed != NULL && newname[0] != '\0') { VERIFY(0 == nvlist_add_boolean(renamed, newname)); } if (error) needagain = B_TRUE; else progress = B_TRUE; } } doagain: fsavl_destroy(local_avl); nvlist_free(local_nv); nvlist_free(deleted); if (needagain && progress) { /* do another pass to fix up temporary names */ if (flags->verbose) (void) printf("another pass:\n"); goto again; } return (needagain); } static int zfs_receive_package(libzfs_handle_t *hdl, int fd, const char *destname, recvflags_t *flags, dmu_replay_record_t *drr, zio_cksum_t *zc, char **top_zfs, int cleanup_fd, uint64_t *action_handlep) { nvlist_t *stream_nv = NULL; avl_tree_t *stream_avl = NULL; char *fromsnap = NULL; char *cp; char tofs[ZFS_MAXNAMELEN]; char sendfs[ZFS_MAXNAMELEN]; char errbuf[1024]; dmu_replay_record_t drre; int error; boolean_t anyerr = B_FALSE; boolean_t softerr = B_FALSE; boolean_t recursive; (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "cannot receive")); assert(drr->drr_type == DRR_BEGIN); assert(drr->drr_u.drr_begin.drr_magic == DMU_BACKUP_MAGIC); assert(DMU_GET_STREAM_HDRTYPE(drr->drr_u.drr_begin.drr_versioninfo) == DMU_COMPOUNDSTREAM); /* * Read in the nvlist from the stream. */ if (drr->drr_payloadlen != 0) { error = recv_read_nvlist(hdl, fd, drr->drr_payloadlen, &stream_nv, flags->byteswap, zc); if (error) { error = zfs_error(hdl, EZFS_BADSTREAM, errbuf); goto out; } } recursive = (nvlist_lookup_boolean(stream_nv, "not_recursive") == ENOENT); if (recursive && strchr(destname, '@')) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "cannot specify snapshot name for multi-snapshot stream")); error = zfs_error(hdl, EZFS_BADSTREAM, errbuf); goto out; } /* * Read in the end record and verify checksum. */ if (0 != (error = recv_read(hdl, fd, &drre, sizeof (drre), flags->byteswap, NULL))) goto out; if (flags->byteswap) { drre.drr_type = BSWAP_32(drre.drr_type); drre.drr_u.drr_end.drr_checksum.zc_word[0] = BSWAP_64(drre.drr_u.drr_end.drr_checksum.zc_word[0]); drre.drr_u.drr_end.drr_checksum.zc_word[1] = BSWAP_64(drre.drr_u.drr_end.drr_checksum.zc_word[1]); drre.drr_u.drr_end.drr_checksum.zc_word[2] = BSWAP_64(drre.drr_u.drr_end.drr_checksum.zc_word[2]); drre.drr_u.drr_end.drr_checksum.zc_word[3] = BSWAP_64(drre.drr_u.drr_end.drr_checksum.zc_word[3]); } if (drre.drr_type != DRR_END) { error = zfs_error(hdl, EZFS_BADSTREAM, errbuf); goto out; } if (!ZIO_CHECKSUM_EQUAL(drre.drr_u.drr_end.drr_checksum, *zc)) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "incorrect header checksum")); error = zfs_error(hdl, EZFS_BADSTREAM, errbuf); goto out; } (void) nvlist_lookup_string(stream_nv, "fromsnap", &fromsnap); if (drr->drr_payloadlen != 0) { nvlist_t *stream_fss; VERIFY(0 == nvlist_lookup_nvlist(stream_nv, "fss", &stream_fss)); if ((stream_avl = fsavl_create(stream_fss)) == NULL) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "couldn't allocate avl tree")); error = zfs_error(hdl, EZFS_NOMEM, errbuf); goto out; } if (fromsnap != NULL) { nvlist_t *renamed = NULL; nvpair_t *pair = NULL; (void) strlcpy(tofs, destname, ZFS_MAXNAMELEN); if (flags->isprefix) { struct drr_begin *drrb = &drr->drr_u.drr_begin; int i; if (flags->istail) { cp = strrchr(drrb->drr_toname, '/'); if (cp == NULL) { (void) strlcat(tofs, "/", ZFS_MAXNAMELEN); i = 0; } else { i = (cp - drrb->drr_toname); } } else { i = strcspn(drrb->drr_toname, "/@"); } /* zfs_receive_one() will create_parents() */ (void) strlcat(tofs, &drrb->drr_toname[i], ZFS_MAXNAMELEN); *strchr(tofs, '@') = '\0'; } if (recursive && !flags->dryrun && !flags->nomount) { VERIFY(0 == nvlist_alloc(&renamed, NV_UNIQUE_NAME, 0)); } softerr = recv_incremental_replication(hdl, tofs, flags, stream_nv, stream_avl, renamed); /* Unmount renamed filesystems before receiving. */ while ((pair = nvlist_next_nvpair(renamed, pair)) != NULL) { zfs_handle_t *zhp; prop_changelist_t *clp = NULL; zhp = zfs_open(hdl, nvpair_name(pair), ZFS_TYPE_FILESYSTEM); if (zhp != NULL) { clp = changelist_gather(zhp, ZFS_PROP_MOUNTPOINT, 0, 0); zfs_close(zhp); if (clp != NULL) { softerr |= changelist_prefix(clp); changelist_free(clp); } } } nvlist_free(renamed); } } /* * Get the fs specified by the first path in the stream (the top level * specified by 'zfs send') and pass it to each invocation of * zfs_receive_one(). */ (void) strlcpy(sendfs, drr->drr_u.drr_begin.drr_toname, ZFS_MAXNAMELEN); if ((cp = strchr(sendfs, '@')) != NULL) *cp = '\0'; /* Finally, receive each contained stream */ do { /* * we should figure out if it has a recoverable * error, in which case do a recv_skip() and drive on. * Note, if we fail due to already having this guid, * zfs_receive_one() will take care of it (ie, * recv_skip() and return 0). */ error = zfs_receive_impl(hdl, destname, flags, fd, sendfs, stream_nv, stream_avl, top_zfs, cleanup_fd, action_handlep); if (error == ENODATA) { error = 0; break; } anyerr |= error; } while (error == 0); if (drr->drr_payloadlen != 0 && fromsnap != NULL) { /* * Now that we have the fs's they sent us, try the * renames again. */ softerr = recv_incremental_replication(hdl, tofs, flags, stream_nv, stream_avl, NULL); } out: fsavl_destroy(stream_avl); if (stream_nv) nvlist_free(stream_nv); if (softerr) error = -2; if (anyerr) error = -1; return (error); } static void trunc_prop_errs(int truncated) { ASSERT(truncated != 0); if (truncated == 1) (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "1 more property could not be set\n")); else (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "%d more properties could not be set\n"), truncated); } static int recv_skip(libzfs_handle_t *hdl, int fd, boolean_t byteswap) { dmu_replay_record_t *drr; void *buf = zfs_alloc(hdl, SPA_MAXBLOCKSIZE); char errbuf[1024]; (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "cannot receive:")); /* XXX would be great to use lseek if possible... */ drr = buf; while (recv_read(hdl, fd, drr, sizeof (dmu_replay_record_t), byteswap, NULL) == 0) { if (byteswap) drr->drr_type = BSWAP_32(drr->drr_type); switch (drr->drr_type) { case DRR_BEGIN: /* NB: not to be used on v2 stream packages */ if (drr->drr_payloadlen != 0) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "invalid substream header")); return (zfs_error(hdl, EZFS_BADSTREAM, errbuf)); } break; case DRR_END: free(buf); return (0); case DRR_OBJECT: if (byteswap) { drr->drr_u.drr_object.drr_bonuslen = BSWAP_32(drr->drr_u.drr_object. drr_bonuslen); } (void) recv_read(hdl, fd, buf, P2ROUNDUP(drr->drr_u.drr_object.drr_bonuslen, 8), B_FALSE, NULL); break; case DRR_WRITE: if (byteswap) { drr->drr_u.drr_write.drr_length = BSWAP_64(drr->drr_u.drr_write.drr_length); } (void) recv_read(hdl, fd, buf, drr->drr_u.drr_write.drr_length, B_FALSE, NULL); break; case DRR_SPILL: if (byteswap) { drr->drr_u.drr_write.drr_length = BSWAP_64(drr->drr_u.drr_spill.drr_length); } (void) recv_read(hdl, fd, buf, drr->drr_u.drr_spill.drr_length, B_FALSE, NULL); break; case DRR_WRITE_EMBEDDED: if (byteswap) { drr->drr_u.drr_write_embedded.drr_psize = BSWAP_32(drr->drr_u.drr_write_embedded. drr_psize); } (void) recv_read(hdl, fd, buf, P2ROUNDUP(drr->drr_u.drr_write_embedded.drr_psize, 8), B_FALSE, NULL); break; case DRR_WRITE_BYREF: case DRR_FREEOBJECTS: case DRR_FREE: break; default: zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "invalid record type")); return (zfs_error(hdl, EZFS_BADSTREAM, errbuf)); } } free(buf); return (-1); } /* * Restores a backup of tosnap from the file descriptor specified by infd. */ static int zfs_receive_one(libzfs_handle_t *hdl, int infd, const char *tosnap, recvflags_t *flags, dmu_replay_record_t *drr, dmu_replay_record_t *drr_noswap, const char *sendfs, nvlist_t *stream_nv, avl_tree_t *stream_avl, char **top_zfs, int cleanup_fd, uint64_t *action_handlep) { zfs_cmd_t zc = { 0 }; time_t begin_time; int ioctl_err, ioctl_errno, err; char *cp; struct drr_begin *drrb = &drr->drr_u.drr_begin; char errbuf[1024]; char prop_errbuf[1024]; const char *chopprefix; boolean_t newfs = B_FALSE; boolean_t stream_wantsnewfs; uint64_t parent_snapguid = 0; prop_changelist_t *clp = NULL; nvlist_t *snapprops_nvlist = NULL; zprop_errflags_t prop_errflags; boolean_t recursive; begin_time = time(NULL); (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "cannot receive")); recursive = (nvlist_lookup_boolean(stream_nv, "not_recursive") == ENOENT); if (stream_avl != NULL) { char *snapname; nvlist_t *fs = fsavl_find(stream_avl, drrb->drr_toguid, &snapname); nvlist_t *props; int ret; (void) nvlist_lookup_uint64(fs, "parentfromsnap", &parent_snapguid); err = nvlist_lookup_nvlist(fs, "props", &props); if (err) VERIFY(0 == nvlist_alloc(&props, NV_UNIQUE_NAME, 0)); if (flags->canmountoff) { VERIFY(0 == nvlist_add_uint64(props, zfs_prop_to_name(ZFS_PROP_CANMOUNT), 0)); } ret = zcmd_write_src_nvlist(hdl, &zc, props); if (err) nvlist_free(props); if (0 == nvlist_lookup_nvlist(fs, "snapprops", &props)) { VERIFY(0 == nvlist_lookup_nvlist(props, snapname, &snapprops_nvlist)); } if (ret != 0) return (-1); } cp = NULL; /* * Determine how much of the snapshot name stored in the stream * we are going to tack on to the name they specified on the * command line, and how much we are going to chop off. * * If they specified a snapshot, chop the entire name stored in * the stream. */ if (flags->istail) { /* * A filesystem was specified with -e. We want to tack on only * the tail of the sent snapshot path. */ if (strchr(tosnap, '@')) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "invalid " "argument - snapshot not allowed with -e")); return (zfs_error(hdl, EZFS_INVALIDNAME, errbuf)); } chopprefix = strrchr(sendfs, '/'); if (chopprefix == NULL) { /* * The tail is the poolname, so we need to * prepend a path separator. */ int len = strlen(drrb->drr_toname); cp = malloc(len + 2); cp[0] = '/'; (void) strcpy(&cp[1], drrb->drr_toname); chopprefix = cp; } else { chopprefix = drrb->drr_toname + (chopprefix - sendfs); } } else if (flags->isprefix) { /* * A filesystem was specified with -d. We want to tack on * everything but the first element of the sent snapshot path * (all but the pool name). */ if (strchr(tosnap, '@')) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "invalid " "argument - snapshot not allowed with -d")); return (zfs_error(hdl, EZFS_INVALIDNAME, errbuf)); } chopprefix = strchr(drrb->drr_toname, '/'); if (chopprefix == NULL) chopprefix = strchr(drrb->drr_toname, '@'); } else if (strchr(tosnap, '@') == NULL) { /* * If a filesystem was specified without -d or -e, we want to * tack on everything after the fs specified by 'zfs send'. */ chopprefix = drrb->drr_toname + strlen(sendfs); } else { /* A snapshot was specified as an exact path (no -d or -e). */ if (recursive) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "cannot specify snapshot name for multi-snapshot " "stream")); return (zfs_error(hdl, EZFS_BADSTREAM, errbuf)); } chopprefix = drrb->drr_toname + strlen(drrb->drr_toname); } ASSERT(strstr(drrb->drr_toname, sendfs) == drrb->drr_toname); ASSERT(chopprefix > drrb->drr_toname); ASSERT(chopprefix <= drrb->drr_toname + strlen(drrb->drr_toname)); ASSERT(chopprefix[0] == '/' || chopprefix[0] == '@' || chopprefix[0] == '\0'); /* * Determine name of destination snapshot, store in zc_value. */ (void) strcpy(zc.zc_value, tosnap); (void) strncat(zc.zc_value, chopprefix, sizeof (zc.zc_value)); #ifdef __FreeBSD__ if (zfs_ioctl_version == ZFS_IOCVER_UNDEF) zfs_ioctl_version = get_zfs_ioctl_version(); /* * For forward compatibility hide tosnap in zc_value */ if (zfs_ioctl_version < ZFS_IOCVER_LZC) (void) strcpy(zc.zc_value + strlen(zc.zc_value) + 1, tosnap); #endif free(cp); if (!zfs_name_valid(zc.zc_value, ZFS_TYPE_SNAPSHOT)) { zcmd_free_nvlists(&zc); return (zfs_error(hdl, EZFS_INVALIDNAME, errbuf)); } /* * Determine the name of the origin snapshot, store in zc_string. */ if (drrb->drr_flags & DRR_FLAG_CLONE) { if (guid_to_name(hdl, zc.zc_value, drrb->drr_fromguid, zc.zc_string) != 0) { zcmd_free_nvlists(&zc); zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "local origin for clone %s does not exist"), zc.zc_value); return (zfs_error(hdl, EZFS_NOENT, errbuf)); } if (flags->verbose) (void) printf("found clone origin %s\n", zc.zc_string); } stream_wantsnewfs = (drrb->drr_fromguid == 0 || (drrb->drr_flags & DRR_FLAG_CLONE)); if (stream_wantsnewfs) { /* * if the parent fs does not exist, look for it based on * the parent snap GUID */ (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "cannot receive new filesystem stream")); (void) strcpy(zc.zc_name, zc.zc_value); cp = strrchr(zc.zc_name, '/'); if (cp) *cp = '\0'; if (cp && !zfs_dataset_exists(hdl, zc.zc_name, ZFS_TYPE_DATASET)) { char suffix[ZFS_MAXNAMELEN]; (void) strcpy(suffix, strrchr(zc.zc_value, '/')); if (guid_to_name(hdl, zc.zc_name, parent_snapguid, zc.zc_value) == 0) { *strchr(zc.zc_value, '@') = '\0'; (void) strcat(zc.zc_value, suffix); } } } else { /* * if the fs does not exist, look for it based on the * fromsnap GUID */ (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "cannot receive incremental stream")); (void) strcpy(zc.zc_name, zc.zc_value); *strchr(zc.zc_name, '@') = '\0'; /* * If the exact receive path was specified and this is the * topmost path in the stream, then if the fs does not exist we * should look no further. */ if ((flags->isprefix || (*(chopprefix = drrb->drr_toname + strlen(sendfs)) != '\0' && *chopprefix != '@')) && !zfs_dataset_exists(hdl, zc.zc_name, ZFS_TYPE_DATASET)) { char snap[ZFS_MAXNAMELEN]; (void) strcpy(snap, strchr(zc.zc_value, '@')); if (guid_to_name(hdl, zc.zc_name, drrb->drr_fromguid, zc.zc_value) == 0) { *strchr(zc.zc_value, '@') = '\0'; (void) strcat(zc.zc_value, snap); } } } (void) strcpy(zc.zc_name, zc.zc_value); *strchr(zc.zc_name, '@') = '\0'; if (zfs_dataset_exists(hdl, zc.zc_name, ZFS_TYPE_DATASET)) { zfs_handle_t *zhp; /* * Destination fs exists. Therefore this should either * be an incremental, or the stream specifies a new fs * (full stream or clone) and they want us to blow it * away (and have therefore specified -F and removed any * snapshots). */ if (stream_wantsnewfs) { if (!flags->force) { zcmd_free_nvlists(&zc); zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "destination '%s' exists\n" "must specify -F to overwrite it"), zc.zc_name); return (zfs_error(hdl, EZFS_EXISTS, errbuf)); } if (ioctl(hdl->libzfs_fd, ZFS_IOC_SNAPSHOT_LIST_NEXT, &zc) == 0) { zcmd_free_nvlists(&zc); zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "destination has snapshots (eg. %s)\n" "must destroy them to overwrite it"), zc.zc_name); return (zfs_error(hdl, EZFS_EXISTS, errbuf)); } } if ((zhp = zfs_open(hdl, zc.zc_name, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME)) == NULL) { zcmd_free_nvlists(&zc); return (-1); } if (stream_wantsnewfs && zhp->zfs_dmustats.dds_origin[0]) { zcmd_free_nvlists(&zc); zfs_close(zhp); zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "destination '%s' is a clone\n" "must destroy it to overwrite it"), zc.zc_name); return (zfs_error(hdl, EZFS_EXISTS, errbuf)); } if (!flags->dryrun && zhp->zfs_type == ZFS_TYPE_FILESYSTEM && stream_wantsnewfs) { /* We can't do online recv in this case */ clp = changelist_gather(zhp, ZFS_PROP_NAME, 0, 0); if (clp == NULL) { zfs_close(zhp); zcmd_free_nvlists(&zc); return (-1); } if (changelist_prefix(clp) != 0) { changelist_free(clp); zfs_close(zhp); zcmd_free_nvlists(&zc); return (-1); } } zfs_close(zhp); } else { /* * Destination filesystem does not exist. Therefore we better * be creating a new filesystem (either from a full backup, or * a clone). It would therefore be invalid if the user * specified only the pool name (i.e. if the destination name * contained no slash character). */ if (!stream_wantsnewfs || (cp = strrchr(zc.zc_name, '/')) == NULL) { zcmd_free_nvlists(&zc); zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "destination '%s' does not exist"), zc.zc_name); return (zfs_error(hdl, EZFS_NOENT, errbuf)); } /* * Trim off the final dataset component so we perform the * recvbackup ioctl to the filesystems's parent. */ *cp = '\0'; if (flags->isprefix && !flags->istail && !flags->dryrun && create_parents(hdl, zc.zc_value, strlen(tosnap)) != 0) { zcmd_free_nvlists(&zc); return (zfs_error(hdl, EZFS_BADRESTORE, errbuf)); } newfs = B_TRUE; } zc.zc_begin_record = drr_noswap->drr_u.drr_begin; zc.zc_cookie = infd; zc.zc_guid = flags->force; if (flags->verbose) { (void) printf("%s %s stream of %s into %s\n", flags->dryrun ? "would receive" : "receiving", drrb->drr_fromguid ? "incremental" : "full", drrb->drr_toname, zc.zc_value); (void) fflush(stdout); } if (flags->dryrun) { zcmd_free_nvlists(&zc); return (recv_skip(hdl, infd, flags->byteswap)); } zc.zc_nvlist_dst = (uint64_t)(uintptr_t)prop_errbuf; zc.zc_nvlist_dst_size = sizeof (prop_errbuf); zc.zc_cleanup_fd = cleanup_fd; zc.zc_action_handle = *action_handlep; err = ioctl_err = zfs_ioctl(hdl, ZFS_IOC_RECV, &zc); ioctl_errno = errno; prop_errflags = (zprop_errflags_t)zc.zc_obj; if (err == 0) { nvlist_t *prop_errors; VERIFY(0 == nvlist_unpack((void *)(uintptr_t)zc.zc_nvlist_dst, zc.zc_nvlist_dst_size, &prop_errors, 0)); nvpair_t *prop_err = NULL; while ((prop_err = nvlist_next_nvpair(prop_errors, prop_err)) != NULL) { char tbuf[1024]; zfs_prop_t prop; int intval; prop = zfs_name_to_prop(nvpair_name(prop_err)); (void) nvpair_value_int32(prop_err, &intval); if (strcmp(nvpair_name(prop_err), ZPROP_N_MORE_ERRORS) == 0) { trunc_prop_errs(intval); break; } else { (void) snprintf(tbuf, sizeof (tbuf), dgettext(TEXT_DOMAIN, "cannot receive %s property on %s"), nvpair_name(prop_err), zc.zc_name); zfs_setprop_error(hdl, prop, intval, tbuf); } } nvlist_free(prop_errors); } zc.zc_nvlist_dst = 0; zc.zc_nvlist_dst_size = 0; zcmd_free_nvlists(&zc); if (err == 0 && snapprops_nvlist) { zfs_cmd_t zc2 = { 0 }; (void) strcpy(zc2.zc_name, zc.zc_value); zc2.zc_cookie = B_TRUE; /* received */ if (zcmd_write_src_nvlist(hdl, &zc2, snapprops_nvlist) == 0) { (void) zfs_ioctl(hdl, ZFS_IOC_SET_PROP, &zc2); zcmd_free_nvlists(&zc2); } } if (err && (ioctl_errno == ENOENT || ioctl_errno == EEXIST)) { /* * It may be that this snapshot already exists, * in which case we want to consume & ignore it * rather than failing. */ avl_tree_t *local_avl; nvlist_t *local_nv, *fs; cp = strchr(zc.zc_value, '@'); /* * XXX Do this faster by just iterating over snaps in * this fs. Also if zc_value does not exist, we will * get a strange "does not exist" error message. */ *cp = '\0'; if (gather_nvlist(hdl, zc.zc_value, NULL, NULL, B_FALSE, &local_nv, &local_avl) == 0) { *cp = '@'; fs = fsavl_find(local_avl, drrb->drr_toguid, NULL); fsavl_destroy(local_avl); nvlist_free(local_nv); if (fs != NULL) { if (flags->verbose) { (void) printf("snap %s already exists; " "ignoring\n", zc.zc_value); } err = ioctl_err = recv_skip(hdl, infd, flags->byteswap); } } *cp = '@'; } if (ioctl_err != 0) { switch (ioctl_errno) { case ENODEV: cp = strchr(zc.zc_value, '@'); *cp = '\0'; zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "most recent snapshot of %s does not\n" "match incremental source"), zc.zc_value); (void) zfs_error(hdl, EZFS_BADRESTORE, errbuf); *cp = '@'; break; case ETXTBSY: zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "destination %s has been modified\n" "since most recent snapshot"), zc.zc_name); (void) zfs_error(hdl, EZFS_BADRESTORE, errbuf); break; case EEXIST: cp = strchr(zc.zc_value, '@'); if (newfs) { /* it's the containing fs that exists */ *cp = '\0'; } zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "destination already exists")); (void) zfs_error_fmt(hdl, EZFS_EXISTS, dgettext(TEXT_DOMAIN, "cannot restore to %s"), zc.zc_value); *cp = '@'; break; case EINVAL: (void) zfs_error(hdl, EZFS_BADSTREAM, errbuf); break; case ECKSUM: zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "invalid stream (checksum mismatch)")); (void) zfs_error(hdl, EZFS_BADSTREAM, errbuf); break; case ENOTSUP: zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "pool must be upgraded to receive this stream.")); (void) zfs_error(hdl, EZFS_BADVERSION, errbuf); break; case EDQUOT: zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "destination %s space quota exceeded"), zc.zc_name); (void) zfs_error(hdl, EZFS_NOSPC, errbuf); break; default: (void) zfs_standard_error(hdl, ioctl_errno, errbuf); } } /* * Mount the target filesystem (if created). Also mount any * children of the target filesystem if we did a replication * receive (indicated by stream_avl being non-NULL). */ cp = strchr(zc.zc_value, '@'); if (cp && (ioctl_err == 0 || !newfs)) { zfs_handle_t *h; *cp = '\0'; h = zfs_open(hdl, zc.zc_value, ZFS_TYPE_FILESYSTEM | ZFS_TYPE_VOLUME); if (h != NULL) { if (h->zfs_type == ZFS_TYPE_VOLUME) { *cp = '@'; } else if (newfs || stream_avl) { /* * Track the first/top of hierarchy fs, * for mounting and sharing later. */ if (top_zfs && *top_zfs == NULL) *top_zfs = zfs_strdup(hdl, zc.zc_value); } zfs_close(h); } *cp = '@'; } if (clp) { err |= changelist_postfix(clp); changelist_free(clp); } if (prop_errflags & ZPROP_ERR_NOCLEAR) { (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "Warning: " "failed to clear unreceived properties on %s"), zc.zc_name); (void) fprintf(stderr, "\n"); } if (prop_errflags & ZPROP_ERR_NORESTORE) { (void) fprintf(stderr, dgettext(TEXT_DOMAIN, "Warning: " "failed to restore original properties on %s"), zc.zc_name); (void) fprintf(stderr, "\n"); } if (err || ioctl_err) return (-1); *action_handlep = zc.zc_action_handle; if (flags->verbose) { char buf1[64]; char buf2[64]; uint64_t bytes = zc.zc_cookie; time_t delta = time(NULL) - begin_time; if (delta == 0) delta = 1; zfs_nicenum(bytes, buf1, sizeof (buf1)); zfs_nicenum(bytes/delta, buf2, sizeof (buf1)); (void) printf("received %sB stream in %lu seconds (%sB/sec)\n", buf1, delta, buf2); } return (0); } static int zfs_receive_impl(libzfs_handle_t *hdl, const char *tosnap, recvflags_t *flags, int infd, const char *sendfs, nvlist_t *stream_nv, avl_tree_t *stream_avl, char **top_zfs, int cleanup_fd, uint64_t *action_handlep) { int err; dmu_replay_record_t drr, drr_noswap; struct drr_begin *drrb = &drr.drr_u.drr_begin; char errbuf[1024]; zio_cksum_t zcksum = { 0 }; uint64_t featureflags; int hdrtype; (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, "cannot receive")); if (flags->isprefix && !zfs_dataset_exists(hdl, tosnap, ZFS_TYPE_DATASET)) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "specified fs " "(%s) does not exist"), tosnap); return (zfs_error(hdl, EZFS_NOENT, errbuf)); } /* read in the BEGIN record */ if (0 != (err = recv_read(hdl, infd, &drr, sizeof (drr), B_FALSE, &zcksum))) return (err); if (drr.drr_type == DRR_END || drr.drr_type == BSWAP_32(DRR_END)) { /* It's the double end record at the end of a package */ return (ENODATA); } /* the kernel needs the non-byteswapped begin record */ drr_noswap = drr; flags->byteswap = B_FALSE; if (drrb->drr_magic == BSWAP_64(DMU_BACKUP_MAGIC)) { /* * We computed the checksum in the wrong byteorder in * recv_read() above; do it again correctly. */ bzero(&zcksum, sizeof (zio_cksum_t)); fletcher_4_incremental_byteswap(&drr, sizeof (drr), &zcksum); flags->byteswap = B_TRUE; drr.drr_type = BSWAP_32(drr.drr_type); drr.drr_payloadlen = BSWAP_32(drr.drr_payloadlen); drrb->drr_magic = BSWAP_64(drrb->drr_magic); drrb->drr_versioninfo = BSWAP_64(drrb->drr_versioninfo); drrb->drr_creation_time = BSWAP_64(drrb->drr_creation_time); drrb->drr_type = BSWAP_32(drrb->drr_type); drrb->drr_flags = BSWAP_32(drrb->drr_flags); drrb->drr_toguid = BSWAP_64(drrb->drr_toguid); drrb->drr_fromguid = BSWAP_64(drrb->drr_fromguid); } if (drrb->drr_magic != DMU_BACKUP_MAGIC || drr.drr_type != DRR_BEGIN) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "invalid " "stream (bad magic number)")); return (zfs_error(hdl, EZFS_BADSTREAM, errbuf)); } featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo); hdrtype = DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo); if (!DMU_STREAM_SUPPORTED(featureflags) || (hdrtype != DMU_SUBSTREAM && hdrtype != DMU_COMPOUNDSTREAM)) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "stream has unsupported feature, feature flags = %lx"), featureflags); return (zfs_error(hdl, EZFS_BADSTREAM, errbuf)); } if (strchr(drrb->drr_toname, '@') == NULL) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "invalid " "stream (bad snapshot name)")); return (zfs_error(hdl, EZFS_BADSTREAM, errbuf)); } if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == DMU_SUBSTREAM) { char nonpackage_sendfs[ZFS_MAXNAMELEN]; if (sendfs == NULL) { /* * We were not called from zfs_receive_package(). Get * the fs specified by 'zfs send'. */ char *cp; (void) strlcpy(nonpackage_sendfs, drr.drr_u.drr_begin.drr_toname, ZFS_MAXNAMELEN); if ((cp = strchr(nonpackage_sendfs, '@')) != NULL) *cp = '\0'; sendfs = nonpackage_sendfs; } return (zfs_receive_one(hdl, infd, tosnap, flags, &drr, &drr_noswap, sendfs, stream_nv, stream_avl, top_zfs, cleanup_fd, action_handlep)); } else { assert(DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == DMU_COMPOUNDSTREAM); return (zfs_receive_package(hdl, infd, tosnap, flags, &drr, &zcksum, top_zfs, cleanup_fd, action_handlep)); } } /* * Restores a backup of tosnap from the file descriptor specified by infd. * Return 0 on total success, -2 if some things couldn't be * destroyed/renamed/promoted, -1 if some things couldn't be received. * (-1 will override -2). */ int zfs_receive(libzfs_handle_t *hdl, const char *tosnap, recvflags_t *flags, int infd, avl_tree_t *stream_avl) { char *top_zfs = NULL; int err; int cleanup_fd; uint64_t action_handle = 0; cleanup_fd = open(ZFS_DEV, O_RDWR|O_EXCL); VERIFY(cleanup_fd >= 0); err = zfs_receive_impl(hdl, tosnap, flags, infd, NULL, NULL, stream_avl, &top_zfs, cleanup_fd, &action_handle); VERIFY(0 == close(cleanup_fd)); if (err == 0 && !flags->nomount && top_zfs) { zfs_handle_t *zhp; prop_changelist_t *clp; zhp = zfs_open(hdl, top_zfs, ZFS_TYPE_FILESYSTEM); if (zhp != NULL) { clp = changelist_gather(zhp, ZFS_PROP_MOUNTPOINT, CL_GATHER_MOUNT_ALWAYS, 0); zfs_close(zhp); if (clp != NULL) { /* mount and share received datasets */ err = changelist_postfix(clp); changelist_free(clp); } } if (zhp == NULL || clp == NULL || err) err = -1; } if (top_zfs) free(top_zfs); return (err); } Index: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_send.c =================================================================== --- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_send.c (revision 286586) +++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_send.c (revision 286587) @@ -1,2236 +1,2306 @@ /* * 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, 2014 by Delphix. All rights reserved. * Copyright (c) 2014, Joyent, Inc. All rights reserved. * Copyright (c) 2012, Martin Matuska . All rights reserved. * Copyright 2014 HybridCluster. All rights reserved. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #ifdef __FreeBSD__ #undef dump_write #define dump_write dmu_dump_write #endif /* Set this tunable to TRUE to replace corrupt data with 0x2f5baddb10c */ int zfs_send_corrupt_data = B_FALSE; static char *dmu_recv_tag = "dmu_recv_tag"; static const char *recv_clone_name = "%recv"; static int dump_bytes(dmu_sendarg_t *dsp, void *buf, int len) { dsl_dataset_t *ds = dsp->dsa_os->os_dsl_dataset; struct uio auio; struct iovec aiov; ASSERT0(len % 8); - fletcher_4_incremental_native(buf, len, &dsp->dsa_zc); aiov.iov_base = buf; aiov.iov_len = len; auio.uio_iov = &aiov; auio.uio_iovcnt = 1; auio.uio_resid = len; auio.uio_segflg = UIO_SYSSPACE; auio.uio_rw = UIO_WRITE; auio.uio_offset = (off_t)-1; auio.uio_td = dsp->dsa_td; #ifdef _KERNEL if (dsp->dsa_fp->f_type == DTYPE_VNODE) bwillwrite(); dsp->dsa_err = fo_write(dsp->dsa_fp, &auio, dsp->dsa_td->td_ucred, 0, dsp->dsa_td); #else fprintf(stderr, "%s: returning EOPNOTSUPP\n", __func__); dsp->dsa_err = EOPNOTSUPP; #endif mutex_enter(&ds->ds_sendstream_lock); *dsp->dsa_off += len; mutex_exit(&ds->ds_sendstream_lock); return (dsp->dsa_err); } +/* + * For all record types except BEGIN, fill in the checksum (overlaid in + * drr_u.drr_checksum.drr_checksum). The checksum verifies everything + * up to the start of the checksum itself. + */ static int +dump_record(dmu_sendarg_t *dsp, void *payload, int payload_len) +{ + ASSERT3U(offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), + ==, sizeof (dmu_replay_record_t) - sizeof (zio_cksum_t)); + fletcher_4_incremental_native(dsp->dsa_drr, + offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), + &dsp->dsa_zc); + if (dsp->dsa_drr->drr_type != DRR_BEGIN) { + ASSERT(ZIO_CHECKSUM_IS_ZERO(&dsp->dsa_drr->drr_u. + drr_checksum.drr_checksum)); + dsp->dsa_drr->drr_u.drr_checksum.drr_checksum = dsp->dsa_zc; + } + fletcher_4_incremental_native(&dsp->dsa_drr-> + drr_u.drr_checksum.drr_checksum, + sizeof (zio_cksum_t), &dsp->dsa_zc); + if (dump_bytes(dsp, dsp->dsa_drr, sizeof (dmu_replay_record_t)) != 0) + return (SET_ERROR(EINTR)); + if (payload_len != 0) { + fletcher_4_incremental_native(payload, payload_len, + &dsp->dsa_zc); + if (dump_bytes(dsp, payload, payload_len) != 0) + return (SET_ERROR(EINTR)); + } + return (0); +} + +static int dump_free(dmu_sendarg_t *dsp, uint64_t object, uint64_t offset, uint64_t length) { struct drr_free *drrf = &(dsp->dsa_drr->drr_u.drr_free); /* * When we receive a free record, dbuf_free_range() assumes * that the receiving system doesn't have any dbufs in the range * being freed. This is always true because there is a one-record * constraint: we only send one WRITE record for any given * object+offset. We know that the one-record constraint is * true because we always send data in increasing order by * object,offset. * * If the increasing-order constraint ever changes, we should find * another way to assert that the one-record constraint is still * satisfied. */ ASSERT(object > dsp->dsa_last_data_object || (object == dsp->dsa_last_data_object && offset > dsp->dsa_last_data_offset)); /* * If we are doing a non-incremental send, then there can't * be any data in the dataset we're receiving into. Therefore * a free record would simply be a no-op. Save space by not * sending it to begin with. */ if (!dsp->dsa_incremental) return (0); if (length != -1ULL && offset + length < offset) length = -1ULL; /* * If there is a pending op, but it's not PENDING_FREE, push it out, * since free block aggregation can only be done for blocks of the * same type (i.e., DRR_FREE records can only be aggregated with * other DRR_FREE records. DRR_FREEOBJECTS records can only be * aggregated with other DRR_FREEOBJECTS records. */ if (dsp->dsa_pending_op != PENDING_NONE && dsp->dsa_pending_op != PENDING_FREE) { - if (dump_bytes(dsp, dsp->dsa_drr, - sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) return (SET_ERROR(EINTR)); dsp->dsa_pending_op = PENDING_NONE; } if (dsp->dsa_pending_op == PENDING_FREE) { /* * There should never be a PENDING_FREE if length is -1 * (because dump_dnode is the only place where this * function is called with a -1, and only after flushing * any pending record). */ ASSERT(length != -1ULL); /* * Check to see whether this free block can be aggregated * with pending one. */ if (drrf->drr_object == object && drrf->drr_offset + drrf->drr_length == offset) { drrf->drr_length += length; return (0); } else { /* not a continuation. Push out pending record */ - if (dump_bytes(dsp, dsp->dsa_drr, - sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) return (SET_ERROR(EINTR)); dsp->dsa_pending_op = PENDING_NONE; } } /* create a FREE record and make it pending */ bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t)); dsp->dsa_drr->drr_type = DRR_FREE; drrf->drr_object = object; drrf->drr_offset = offset; drrf->drr_length = length; drrf->drr_toguid = dsp->dsa_toguid; if (length == -1ULL) { - if (dump_bytes(dsp, dsp->dsa_drr, - sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) return (SET_ERROR(EINTR)); } else { dsp->dsa_pending_op = PENDING_FREE; } return (0); } static int dump_write(dmu_sendarg_t *dsp, dmu_object_type_t type, uint64_t object, uint64_t offset, int blksz, const blkptr_t *bp, void *data) { struct drr_write *drrw = &(dsp->dsa_drr->drr_u.drr_write); /* * We send data in increasing object, offset order. * See comment in dump_free() for details. */ ASSERT(object > dsp->dsa_last_data_object || (object == dsp->dsa_last_data_object && offset > dsp->dsa_last_data_offset)); dsp->dsa_last_data_object = object; dsp->dsa_last_data_offset = offset + blksz - 1; /* * If there is any kind of pending aggregation (currently either * a grouping of free objects or free blocks), push it out to * the stream, since aggregation can't be done across operations * of different types. */ if (dsp->dsa_pending_op != PENDING_NONE) { - if (dump_bytes(dsp, dsp->dsa_drr, - sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) return (SET_ERROR(EINTR)); dsp->dsa_pending_op = PENDING_NONE; } - /* write a DATA record */ + /* write a WRITE record */ bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t)); dsp->dsa_drr->drr_type = DRR_WRITE; drrw->drr_object = object; drrw->drr_type = type; drrw->drr_offset = offset; drrw->drr_length = blksz; drrw->drr_toguid = dsp->dsa_toguid; if (bp == NULL || BP_IS_EMBEDDED(bp)) { /* * There's no pre-computed checksum for partial-block * writes or embedded BP's, so (like * fletcher4-checkummed blocks) userland will have to * compute a dedup-capable checksum itself. */ drrw->drr_checksumtype = ZIO_CHECKSUM_OFF; } else { drrw->drr_checksumtype = BP_GET_CHECKSUM(bp); if (zio_checksum_table[drrw->drr_checksumtype].ci_dedup) drrw->drr_checksumflags |= DRR_CHECKSUM_DEDUP; DDK_SET_LSIZE(&drrw->drr_key, BP_GET_LSIZE(bp)); DDK_SET_PSIZE(&drrw->drr_key, BP_GET_PSIZE(bp)); DDK_SET_COMPRESS(&drrw->drr_key, BP_GET_COMPRESS(bp)); drrw->drr_key.ddk_cksum = bp->blk_cksum; } - if (dump_bytes(dsp, dsp->dsa_drr, sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, data, blksz) != 0) return (SET_ERROR(EINTR)); - if (dump_bytes(dsp, data, blksz) != 0) - return (SET_ERROR(EINTR)); return (0); } static int dump_write_embedded(dmu_sendarg_t *dsp, uint64_t object, uint64_t offset, int blksz, const blkptr_t *bp) { char buf[BPE_PAYLOAD_SIZE]; struct drr_write_embedded *drrw = &(dsp->dsa_drr->drr_u.drr_write_embedded); if (dsp->dsa_pending_op != PENDING_NONE) { - if (dump_bytes(dsp, dsp->dsa_drr, - sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) return (EINTR); dsp->dsa_pending_op = PENDING_NONE; } ASSERT(BP_IS_EMBEDDED(bp)); bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t)); dsp->dsa_drr->drr_type = DRR_WRITE_EMBEDDED; drrw->drr_object = object; drrw->drr_offset = offset; drrw->drr_length = blksz; drrw->drr_toguid = dsp->dsa_toguid; drrw->drr_compression = BP_GET_COMPRESS(bp); drrw->drr_etype = BPE_GET_ETYPE(bp); drrw->drr_lsize = BPE_GET_LSIZE(bp); drrw->drr_psize = BPE_GET_PSIZE(bp); decode_embedded_bp_compressed(bp, buf); - if (dump_bytes(dsp, dsp->dsa_drr, sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, buf, P2ROUNDUP(drrw->drr_psize, 8)) != 0) return (EINTR); - if (dump_bytes(dsp, buf, P2ROUNDUP(drrw->drr_psize, 8)) != 0) - return (EINTR); return (0); } static int dump_spill(dmu_sendarg_t *dsp, uint64_t object, int blksz, void *data) { struct drr_spill *drrs = &(dsp->dsa_drr->drr_u.drr_spill); if (dsp->dsa_pending_op != PENDING_NONE) { - if (dump_bytes(dsp, dsp->dsa_drr, - sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) return (SET_ERROR(EINTR)); dsp->dsa_pending_op = PENDING_NONE; } /* write a SPILL record */ bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t)); dsp->dsa_drr->drr_type = DRR_SPILL; drrs->drr_object = object; drrs->drr_length = blksz; drrs->drr_toguid = dsp->dsa_toguid; - if (dump_bytes(dsp, dsp->dsa_drr, sizeof (dmu_replay_record_t))) + if (dump_record(dsp, data, blksz) != 0) return (SET_ERROR(EINTR)); - if (dump_bytes(dsp, data, blksz)) - return (SET_ERROR(EINTR)); return (0); } static int dump_freeobjects(dmu_sendarg_t *dsp, uint64_t firstobj, uint64_t numobjs) { struct drr_freeobjects *drrfo = &(dsp->dsa_drr->drr_u.drr_freeobjects); /* See comment in dump_free(). */ if (!dsp->dsa_incremental) return (0); /* * If there is a pending op, but it's not PENDING_FREEOBJECTS, * push it out, since free block aggregation can only be done for * blocks of the same type (i.e., DRR_FREE records can only be * aggregated with other DRR_FREE records. DRR_FREEOBJECTS records * can only be aggregated with other DRR_FREEOBJECTS records. */ if (dsp->dsa_pending_op != PENDING_NONE && dsp->dsa_pending_op != PENDING_FREEOBJECTS) { - if (dump_bytes(dsp, dsp->dsa_drr, - sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) return (SET_ERROR(EINTR)); dsp->dsa_pending_op = PENDING_NONE; } if (dsp->dsa_pending_op == PENDING_FREEOBJECTS) { /* * See whether this free object array can be aggregated * with pending one */ if (drrfo->drr_firstobj + drrfo->drr_numobjs == firstobj) { drrfo->drr_numobjs += numobjs; return (0); } else { /* can't be aggregated. Push out pending record */ - if (dump_bytes(dsp, dsp->dsa_drr, - sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) return (SET_ERROR(EINTR)); dsp->dsa_pending_op = PENDING_NONE; } } /* write a FREEOBJECTS record */ bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t)); dsp->dsa_drr->drr_type = DRR_FREEOBJECTS; drrfo->drr_firstobj = firstobj; drrfo->drr_numobjs = numobjs; drrfo->drr_toguid = dsp->dsa_toguid; dsp->dsa_pending_op = PENDING_FREEOBJECTS; return (0); } static int dump_dnode(dmu_sendarg_t *dsp, uint64_t object, dnode_phys_t *dnp) { struct drr_object *drro = &(dsp->dsa_drr->drr_u.drr_object); if (dnp == NULL || dnp->dn_type == DMU_OT_NONE) return (dump_freeobjects(dsp, object, 1)); if (dsp->dsa_pending_op != PENDING_NONE) { - if (dump_bytes(dsp, dsp->dsa_drr, - sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) return (SET_ERROR(EINTR)); dsp->dsa_pending_op = PENDING_NONE; } /* write an OBJECT record */ bzero(dsp->dsa_drr, sizeof (dmu_replay_record_t)); dsp->dsa_drr->drr_type = DRR_OBJECT; drro->drr_object = object; drro->drr_type = dnp->dn_type; drro->drr_bonustype = dnp->dn_bonustype; drro->drr_blksz = dnp->dn_datablkszsec << SPA_MINBLOCKSHIFT; drro->drr_bonuslen = dnp->dn_bonuslen; drro->drr_checksumtype = dnp->dn_checksum; drro->drr_compress = dnp->dn_compress; drro->drr_toguid = dsp->dsa_toguid; if (!(dsp->dsa_featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) && drro->drr_blksz > SPA_OLD_MAXBLOCKSIZE) drro->drr_blksz = SPA_OLD_MAXBLOCKSIZE; - if (dump_bytes(dsp, dsp->dsa_drr, sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, DN_BONUS(dnp), + P2ROUNDUP(dnp->dn_bonuslen, 8)) != 0) { return (SET_ERROR(EINTR)); + } - if (dump_bytes(dsp, DN_BONUS(dnp), P2ROUNDUP(dnp->dn_bonuslen, 8)) != 0) - return (SET_ERROR(EINTR)); - /* Free anything past the end of the file. */ if (dump_free(dsp, object, (dnp->dn_maxblkid + 1) * (dnp->dn_datablkszsec << SPA_MINBLOCKSHIFT), -1ULL) != 0) return (SET_ERROR(EINTR)); if (dsp->dsa_err != 0) return (SET_ERROR(EINTR)); return (0); } static boolean_t backup_do_embed(dmu_sendarg_t *dsp, const blkptr_t *bp) { if (!BP_IS_EMBEDDED(bp)) return (B_FALSE); /* * Compression function must be legacy, or explicitly enabled. */ if ((BP_GET_COMPRESS(bp) >= ZIO_COMPRESS_LEGACY_FUNCTIONS && !(dsp->dsa_featureflags & DMU_BACKUP_FEATURE_EMBED_DATA_LZ4))) return (B_FALSE); /* * Embed type must be explicitly enabled. */ switch (BPE_GET_ETYPE(bp)) { case BP_EMBEDDED_TYPE_DATA: if (dsp->dsa_featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) return (B_TRUE); break; default: return (B_FALSE); } return (B_FALSE); } #define BP_SPAN(dnp, level) \ (((uint64_t)dnp->dn_datablkszsec) << (SPA_MINBLOCKSHIFT + \ (level) * (dnp->dn_indblkshift - SPA_BLKPTRSHIFT))) /* ARGSUSED */ static int backup_cb(spa_t *spa, zilog_t *zilog, const blkptr_t *bp, const zbookmark_phys_t *zb, const dnode_phys_t *dnp, void *arg) { dmu_sendarg_t *dsp = arg; dmu_object_type_t type = bp ? BP_GET_TYPE(bp) : DMU_OT_NONE; int err = 0; if (issig(JUSTLOOKING) && issig(FORREAL)) return (SET_ERROR(EINTR)); if (zb->zb_object != DMU_META_DNODE_OBJECT && DMU_OBJECT_IS_SPECIAL(zb->zb_object)) { return (0); } else if (zb->zb_level == ZB_ZIL_LEVEL) { /* * If we are sending a non-snapshot (which is allowed on * read-only pools), it may have a ZIL, which must be ignored. */ return (0); } else if (BP_IS_HOLE(bp) && zb->zb_object == DMU_META_DNODE_OBJECT) { uint64_t span = BP_SPAN(dnp, zb->zb_level); uint64_t dnobj = (zb->zb_blkid * span) >> DNODE_SHIFT; err = dump_freeobjects(dsp, dnobj, span >> DNODE_SHIFT); } else if (BP_IS_HOLE(bp)) { uint64_t span = BP_SPAN(dnp, zb->zb_level); err = dump_free(dsp, zb->zb_object, zb->zb_blkid * span, span); } else if (zb->zb_level > 0 || type == DMU_OT_OBJSET) { return (0); } else if (type == DMU_OT_DNODE) { dnode_phys_t *blk; int i; int blksz = BP_GET_LSIZE(bp); arc_flags_t aflags = ARC_FLAG_WAIT; arc_buf_t *abuf; if (arc_read(NULL, spa, bp, arc_getbuf_func, &abuf, ZIO_PRIORITY_ASYNC_READ, ZIO_FLAG_CANFAIL, &aflags, zb) != 0) return (SET_ERROR(EIO)); blk = abuf->b_data; for (i = 0; i < blksz >> DNODE_SHIFT; i++) { uint64_t dnobj = (zb->zb_blkid << (DNODE_BLOCK_SHIFT - DNODE_SHIFT)) + i; err = dump_dnode(dsp, dnobj, blk+i); if (err != 0) break; } (void) arc_buf_remove_ref(abuf, &abuf); } else if (type == DMU_OT_SA) { arc_flags_t aflags = ARC_FLAG_WAIT; arc_buf_t *abuf; int blksz = BP_GET_LSIZE(bp); if (arc_read(NULL, spa, bp, arc_getbuf_func, &abuf, ZIO_PRIORITY_ASYNC_READ, ZIO_FLAG_CANFAIL, &aflags, zb) != 0) return (SET_ERROR(EIO)); err = dump_spill(dsp, zb->zb_object, blksz, abuf->b_data); (void) arc_buf_remove_ref(abuf, &abuf); } else if (backup_do_embed(dsp, bp)) { /* it's an embedded level-0 block of a regular object */ int blksz = dnp->dn_datablkszsec << SPA_MINBLOCKSHIFT; err = dump_write_embedded(dsp, zb->zb_object, zb->zb_blkid * blksz, blksz, bp); } else { /* it's a level-0 block of a regular object */ arc_flags_t aflags = ARC_FLAG_WAIT; arc_buf_t *abuf; int blksz = BP_GET_LSIZE(bp); uint64_t offset; ASSERT3U(blksz, ==, dnp->dn_datablkszsec << SPA_MINBLOCKSHIFT); ASSERT0(zb->zb_level); if (arc_read(NULL, spa, bp, arc_getbuf_func, &abuf, ZIO_PRIORITY_ASYNC_READ, ZIO_FLAG_CANFAIL, &aflags, zb) != 0) { if (zfs_send_corrupt_data) { /* Send a block filled with 0x"zfs badd bloc" */ abuf = arc_buf_alloc(spa, blksz, &abuf, ARC_BUFC_DATA); uint64_t *ptr; for (ptr = abuf->b_data; (char *)ptr < (char *)abuf->b_data + blksz; ptr++) *ptr = 0x2f5baddb10cULL; } else { return (SET_ERROR(EIO)); } } offset = zb->zb_blkid * blksz; if (!(dsp->dsa_featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) && blksz > SPA_OLD_MAXBLOCKSIZE) { char *buf = abuf->b_data; while (blksz > 0 && err == 0) { int n = MIN(blksz, SPA_OLD_MAXBLOCKSIZE); err = dump_write(dsp, type, zb->zb_object, offset, n, NULL, buf); offset += n; buf += n; blksz -= n; } } else { err = dump_write(dsp, type, zb->zb_object, offset, blksz, bp, abuf->b_data); } (void) arc_buf_remove_ref(abuf, &abuf); } ASSERT(err == 0 || err == EINTR); return (err); } /* * Releases dp using the specified tag. */ static int dmu_send_impl(void *tag, dsl_pool_t *dp, dsl_dataset_t *ds, zfs_bookmark_phys_t *fromzb, boolean_t is_clone, boolean_t embedok, #ifdef illumos boolean_t large_block_ok, int outfd, vnode_t *vp, offset_t *off) #else boolean_t large_block_ok, int outfd, struct file *fp, offset_t *off) #endif { objset_t *os; dmu_replay_record_t *drr; dmu_sendarg_t *dsp; int err; uint64_t fromtxg = 0; uint64_t featureflags = 0; err = dmu_objset_from_ds(ds, &os); if (err != 0) { dsl_pool_rele(dp, tag); return (err); } drr = kmem_zalloc(sizeof (dmu_replay_record_t), KM_SLEEP); drr->drr_type = DRR_BEGIN; drr->drr_u.drr_begin.drr_magic = DMU_BACKUP_MAGIC; DMU_SET_STREAM_HDRTYPE(drr->drr_u.drr_begin.drr_versioninfo, DMU_SUBSTREAM); #ifdef _KERNEL if (dmu_objset_type(os) == DMU_OST_ZFS) { uint64_t version; if (zfs_get_zplprop(os, ZFS_PROP_VERSION, &version) != 0) { kmem_free(drr, sizeof (dmu_replay_record_t)); dsl_pool_rele(dp, tag); return (SET_ERROR(EINVAL)); } if (version >= ZPL_VERSION_SA) { featureflags |= DMU_BACKUP_FEATURE_SA_SPILL; } } #endif if (large_block_ok && ds->ds_large_blocks) featureflags |= DMU_BACKUP_FEATURE_LARGE_BLOCKS; if (embedok && spa_feature_is_active(dp->dp_spa, SPA_FEATURE_EMBEDDED_DATA)) { featureflags |= DMU_BACKUP_FEATURE_EMBED_DATA; if (spa_feature_is_active(dp->dp_spa, SPA_FEATURE_LZ4_COMPRESS)) featureflags |= DMU_BACKUP_FEATURE_EMBED_DATA_LZ4; } else { embedok = B_FALSE; } DMU_SET_FEATUREFLAGS(drr->drr_u.drr_begin.drr_versioninfo, featureflags); drr->drr_u.drr_begin.drr_creation_time = dsl_dataset_phys(ds)->ds_creation_time; drr->drr_u.drr_begin.drr_type = dmu_objset_type(os); if (is_clone) drr->drr_u.drr_begin.drr_flags |= DRR_FLAG_CLONE; drr->drr_u.drr_begin.drr_toguid = dsl_dataset_phys(ds)->ds_guid; if (dsl_dataset_phys(ds)->ds_flags & DS_FLAG_CI_DATASET) drr->drr_u.drr_begin.drr_flags |= DRR_FLAG_CI_DATA; if (fromzb != NULL) { drr->drr_u.drr_begin.drr_fromguid = fromzb->zbm_guid; fromtxg = fromzb->zbm_creation_txg; } dsl_dataset_name(ds, drr->drr_u.drr_begin.drr_toname); if (!ds->ds_is_snapshot) { (void) strlcat(drr->drr_u.drr_begin.drr_toname, "@--head--", sizeof (drr->drr_u.drr_begin.drr_toname)); } dsp = kmem_zalloc(sizeof (dmu_sendarg_t), KM_SLEEP); dsp->dsa_drr = drr; dsp->dsa_outfd = outfd; dsp->dsa_proc = curproc; dsp->dsa_td = curthread; dsp->dsa_fp = fp; dsp->dsa_os = os; dsp->dsa_off = off; dsp->dsa_toguid = dsl_dataset_phys(ds)->ds_guid; - ZIO_SET_CHECKSUM(&dsp->dsa_zc, 0, 0, 0, 0); dsp->dsa_pending_op = PENDING_NONE; dsp->dsa_incremental = (fromzb != NULL); dsp->dsa_featureflags = featureflags; mutex_enter(&ds->ds_sendstream_lock); list_insert_head(&ds->ds_sendstreams, dsp); mutex_exit(&ds->ds_sendstream_lock); dsl_dataset_long_hold(ds, FTAG); dsl_pool_rele(dp, tag); - if (dump_bytes(dsp, drr, sizeof (dmu_replay_record_t)) != 0) { + if (dump_record(dsp, NULL, 0) != 0) { err = dsp->dsa_err; goto out; } err = traverse_dataset(ds, fromtxg, TRAVERSE_PRE | TRAVERSE_PREFETCH, backup_cb, dsp); if (dsp->dsa_pending_op != PENDING_NONE) - if (dump_bytes(dsp, drr, sizeof (dmu_replay_record_t)) != 0) + if (dump_record(dsp, NULL, 0) != 0) err = SET_ERROR(EINTR); if (err != 0) { if (err == EINTR && dsp->dsa_err != 0) err = dsp->dsa_err; goto out; } bzero(drr, sizeof (dmu_replay_record_t)); drr->drr_type = DRR_END; drr->drr_u.drr_end.drr_checksum = dsp->dsa_zc; drr->drr_u.drr_end.drr_toguid = dsp->dsa_toguid; - if (dump_bytes(dsp, drr, sizeof (dmu_replay_record_t)) != 0) { + if (dump_record(dsp, NULL, 0) != 0) { err = dsp->dsa_err; goto out; } out: mutex_enter(&ds->ds_sendstream_lock); list_remove(&ds->ds_sendstreams, dsp); mutex_exit(&ds->ds_sendstream_lock); kmem_free(drr, sizeof (dmu_replay_record_t)); kmem_free(dsp, sizeof (dmu_sendarg_t)); dsl_dataset_long_rele(ds, FTAG); return (err); } int dmu_send_obj(const char *pool, uint64_t tosnap, uint64_t fromsnap, boolean_t embedok, boolean_t large_block_ok, #ifdef illumos int outfd, vnode_t *vp, offset_t *off) #else int outfd, struct file *fp, offset_t *off) #endif { dsl_pool_t *dp; dsl_dataset_t *ds; dsl_dataset_t *fromds = NULL; int err; err = dsl_pool_hold(pool, FTAG, &dp); if (err != 0) return (err); err = dsl_dataset_hold_obj(dp, tosnap, FTAG, &ds); if (err != 0) { dsl_pool_rele(dp, FTAG); return (err); } if (fromsnap != 0) { zfs_bookmark_phys_t zb; boolean_t is_clone; err = dsl_dataset_hold_obj(dp, fromsnap, FTAG, &fromds); if (err != 0) { dsl_dataset_rele(ds, FTAG); dsl_pool_rele(dp, FTAG); return (err); } if (!dsl_dataset_is_before(ds, fromds, 0)) err = SET_ERROR(EXDEV); zb.zbm_creation_time = dsl_dataset_phys(fromds)->ds_creation_time; zb.zbm_creation_txg = dsl_dataset_phys(fromds)->ds_creation_txg; zb.zbm_guid = dsl_dataset_phys(fromds)->ds_guid; is_clone = (fromds->ds_dir != ds->ds_dir); dsl_dataset_rele(fromds, FTAG); err = dmu_send_impl(FTAG, dp, ds, &zb, is_clone, embedok, large_block_ok, outfd, fp, off); } else { err = dmu_send_impl(FTAG, dp, ds, NULL, B_FALSE, embedok, large_block_ok, outfd, fp, off); } dsl_dataset_rele(ds, FTAG); return (err); } int dmu_send(const char *tosnap, const char *fromsnap, boolean_t embedok, boolean_t large_block_ok, #ifdef illumos int outfd, vnode_t *vp, offset_t *off) #else int outfd, struct file *fp, offset_t *off) #endif { dsl_pool_t *dp; dsl_dataset_t *ds; int err; boolean_t owned = B_FALSE; if (fromsnap != NULL && strpbrk(fromsnap, "@#") == NULL) return (SET_ERROR(EINVAL)); err = dsl_pool_hold(tosnap, FTAG, &dp); if (err != 0) return (err); if (strchr(tosnap, '@') == NULL && spa_writeable(dp->dp_spa)) { /* * We are sending a filesystem or volume. Ensure * that it doesn't change by owning the dataset. */ err = dsl_dataset_own(dp, tosnap, FTAG, &ds); owned = B_TRUE; } else { err = dsl_dataset_hold(dp, tosnap, FTAG, &ds); } if (err != 0) { dsl_pool_rele(dp, FTAG); return (err); } if (fromsnap != NULL) { zfs_bookmark_phys_t zb; boolean_t is_clone = B_FALSE; int fsnamelen = strchr(tosnap, '@') - tosnap; /* * If the fromsnap is in a different filesystem, then * mark the send stream as a clone. */ if (strncmp(tosnap, fromsnap, fsnamelen) != 0 || (fromsnap[fsnamelen] != '@' && fromsnap[fsnamelen] != '#')) { is_clone = B_TRUE; } if (strchr(fromsnap, '@')) { dsl_dataset_t *fromds; err = dsl_dataset_hold(dp, fromsnap, FTAG, &fromds); if (err == 0) { if (!dsl_dataset_is_before(ds, fromds, 0)) err = SET_ERROR(EXDEV); zb.zbm_creation_time = dsl_dataset_phys(fromds)->ds_creation_time; zb.zbm_creation_txg = dsl_dataset_phys(fromds)->ds_creation_txg; zb.zbm_guid = dsl_dataset_phys(fromds)->ds_guid; is_clone = (ds->ds_dir != fromds->ds_dir); dsl_dataset_rele(fromds, FTAG); } } else { err = dsl_bookmark_lookup(dp, fromsnap, ds, &zb); } if (err != 0) { dsl_dataset_rele(ds, FTAG); dsl_pool_rele(dp, FTAG); return (err); } err = dmu_send_impl(FTAG, dp, ds, &zb, is_clone, embedok, large_block_ok, outfd, fp, off); } else { err = dmu_send_impl(FTAG, dp, ds, NULL, B_FALSE, embedok, large_block_ok, outfd, fp, off); } if (owned) dsl_dataset_disown(ds, FTAG); else dsl_dataset_rele(ds, FTAG); return (err); } int dmu_send_estimate(dsl_dataset_t *ds, dsl_dataset_t *fromds, uint64_t *sizep) { dsl_pool_t *dp = ds->ds_dir->dd_pool; int err; uint64_t size; ASSERT(dsl_pool_config_held(dp)); /* tosnap must be a snapshot */ if (!ds->ds_is_snapshot) return (SET_ERROR(EINVAL)); /* fromsnap, if provided, must be a snapshot */ if (fromds != NULL && !fromds->ds_is_snapshot) return (SET_ERROR(EINVAL)); /* * fromsnap must be an earlier snapshot from the same fs as tosnap, * or the origin's fs. */ if (fromds != NULL && !dsl_dataset_is_before(ds, fromds, 0)) return (SET_ERROR(EXDEV)); /* Get uncompressed size estimate of changed data. */ if (fromds == NULL) { size = dsl_dataset_phys(ds)->ds_uncompressed_bytes; } else { uint64_t used, comp; err = dsl_dataset_space_written(fromds, ds, &used, &comp, &size); if (err != 0) return (err); } /* * Assume that space (both on-disk and in-stream) is dominated by * data. We will adjust for indirect blocks and the copies property, * but ignore per-object space used (eg, dnodes and DRR_OBJECT records). */ /* * Subtract out approximate space used by indirect blocks. * Assume most space is used by data blocks (non-indirect, non-dnode). * Assume all blocks are recordsize. Assume ditto blocks and * internal fragmentation counter out compression. * * Therefore, space used by indirect blocks is sizeof(blkptr_t) per * block, which we observe in practice. */ uint64_t recordsize; err = dsl_prop_get_int_ds(ds, "recordsize", &recordsize); if (err != 0) return (err); size -= size / recordsize * sizeof (blkptr_t); /* Add in the space for the record associated with each block. */ size += size / recordsize * sizeof (dmu_replay_record_t); *sizep = size; return (0); } 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); 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)); /* * The receiving code doesn't know how to translate a WRITE_EMBEDDED * record to a plan 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_EMBED_DATA_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. */ if ((featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) && !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LARGE_BLOCKS)) 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) { 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[MAXNAMELEN]; /* * If it's a non-clone incremental, we are missing the * target fs, so fail the recv. */ if (fromguid != 0 && !(flags & DRR_FLAG_CLONE)) return (SET_ERROR(ENOENT)); /* Open the parent of tofs */ ASSERT3U(strlen(tofs), <, MAXNAMELEN); (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) { 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); 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; crflags = (drrb->drr_flags & DRR_FLAG_CI_DATA) ? DS_FLAG_CI_DATASET : 0; 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 ((DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo) & DMU_BACKUP_FEATURE_LARGE_BLOCKS) && !newds->ds_large_blocks) { dsl_dataset_activate_large_blocks_sync_impl(dsobj, tx); newds->ds_large_blocks = B_TRUE; } 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. */ 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); } drba->drba_cookie->drc_ds = newds; spa_history_log_internal_ds(newds, "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, struct drr_begin *drrb, boolean_t force, char *origin, dmu_recv_cookie_t *drc) { dmu_recv_begin_arg_t drba = { 0 }; dmu_replay_record_t *drr; bzero(drc, sizeof (dmu_recv_cookie_t)); drc->drc_drrb = drrb; drc->drc_tosnap = tosnap; drc->drc_tofs = tofs; drc->drc_force = force; drc->drc_cred = CRED(); if (drrb->drr_magic == BSWAP_64(DMU_BACKUP_MAGIC)) drc->drc_byteswap = B_TRUE; else if (drrb->drr_magic != DMU_BACKUP_MAGIC) return (SET_ERROR(EINVAL)); drr = kmem_zalloc(sizeof (dmu_replay_record_t), KM_SLEEP); drr->drr_type = DRR_BEGIN; drr->drr_u.drr_begin = *drc->drc_drrb; if (drc->drc_byteswap) { fletcher_4_incremental_byteswap(drr, sizeof (dmu_replay_record_t), &drc->drc_cksum); } else { fletcher_4_incremental_native(drr, sizeof (dmu_replay_record_t), &drc->drc_cksum); } kmem_free(drr, sizeof (dmu_replay_record_t)); if (drc->drc_byteswap) { drrb->drr_magic = BSWAP_64(drrb->drr_magic); drrb->drr_versioninfo = BSWAP_64(drrb->drr_versioninfo); drrb->drr_creation_time = BSWAP_64(drrb->drr_creation_time); drrb->drr_type = BSWAP_32(drrb->drr_type); drrb->drr_toguid = BSWAP_64(drrb->drr_toguid); drrb->drr_fromguid = BSWAP_64(drrb->drr_fromguid); } drba.drba_origin = origin; drba.drba_cookie = drc; drba.drba_cred = CRED(); return (dsl_sync_task(tofs, dmu_recv_begin_check, dmu_recv_begin_sync, &drba, 5, ZFS_SPACE_CHECK_NORMAL)); } struct restorearg { + objset_t *os; int err; boolean_t byteswap; kthread_t *td; struct file *fp; - char *buf; uint64_t voff; int bufsize; /* amount of memory allocated for buf */ + + dmu_replay_record_t *drr; + dmu_replay_record_t *next_drr; + char *buf; zio_cksum_t cksum; + zio_cksum_t prev_cksum; + avl_tree_t *guid_to_ds_map; }; 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 restore_bytes(struct restorearg *ra, void *buf, int len, off_t off, ssize_t *resid) { struct uio auio; struct iovec aiov; int error; aiov.iov_base = buf; aiov.iov_len = len; auio.uio_iov = &aiov; auio.uio_iovcnt = 1; auio.uio_resid = len; auio.uio_segflg = UIO_SYSSPACE; auio.uio_rw = UIO_READ; auio.uio_offset = off; auio.uio_td = ra->td; #ifdef _KERNEL error = fo_read(ra->fp, &auio, ra->td->td_ucred, FOF_OFFSET, ra->td); #else fprintf(stderr, "%s: returning EOPNOTSUPP\n", __func__); error = EOPNOTSUPP; #endif *resid = auio.uio_resid; return (error); } -static void * -restore_read(struct restorearg *ra, int len, char *buf) +static int +restore_read(struct restorearg *ra, int len, void *buf) { int done = 0; - if (buf == NULL) - buf = ra->buf; - /* some things will require 8-byte alignment, so everything must */ ASSERT0(len % 8); ASSERT3U(len, <=, ra->bufsize); while (done < len) { ssize_t resid; ra->err = restore_bytes(ra, buf + done, len - done, ra->voff, &resid); if (resid == len - done) ra->err = SET_ERROR(EINVAL); ra->voff += len - done - resid; done = len - resid; if (ra->err != 0) - return (NULL); + return (ra->err); } ASSERT3U(done, ==, len); - if (ra->byteswap) - fletcher_4_incremental_byteswap(buf, len, &ra->cksum); - else - fletcher_4_incremental_native(buf, len, &ra->cksum); - return (buf); + return (0); } static void -backup_byteswap(dmu_replay_record_t *drr) +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_length); DO64(drr_write.drr_toguid); - DO64(drr_write.drr_key.ddk_cksum.zc_word[0]); - DO64(drr_write.drr_key.ddk_cksum.zc_word[1]); - DO64(drr_write.drr_key.ddk_cksum.zc_word[2]); - DO64(drr_write.drr_key.ddk_cksum.zc_word[3]); + ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_write.drr_key.ddk_cksum); DO64(drr_write.drr_key.ddk_prop); 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); - DO64(drr_write_byref.drr_key.ddk_cksum.zc_word[0]); - DO64(drr_write_byref.drr_key.ddk_cksum.zc_word[1]); - DO64(drr_write_byref.drr_key.ddk_cksum.zc_word[2]); - DO64(drr_write_byref.drr_key.ddk_cksum.zc_word[3]); + 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_checksum.zc_word[0]); - DO64(drr_end.drr_checksum.zc_word[1]); - DO64(drr_end.drr_checksum.zc_word[2]); - DO64(drr_end.drr_checksum.zc_word[3]); 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_MAX_BONUSLEN - bonus_size) >> SPA_BLKPTRSHIFT)); } } static int -restore_object(struct restorearg *ra, objset_t *os, struct drr_object *drro) +restore_object(struct restorearg *ra, struct drr_object *drro, void *data) { dmu_object_info_t doi; dmu_tx_t *tx; - void *data = NULL; uint64_t object; int err; 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(os)) || + drro->drr_blksz > spa_maxblocksize(dmu_objset_spa(ra->os)) || drro->drr_bonuslen > DN_MAX_BONUSLEN) { return (SET_ERROR(EINVAL)); } - err = dmu_object_info(os, drro->drr_object, &doi); + err = dmu_object_info(ra->os, drro->drr_object, &doi); if (err != 0 && err != ENOENT) return (SET_ERROR(EINVAL)); object = err == 0 ? drro->drr_object : DMU_NEW_OBJECT; - if (drro->drr_bonuslen) { - data = restore_read(ra, P2ROUNDUP(drro->drr_bonuslen, 8), NULL); - if (ra->err != 0) - return (ra->err); - } - /* * 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; nblkptr = deduce_nblkptr(drro->drr_bonustype, drro->drr_bonuslen); if (drro->drr_blksz != doi.doi_data_block_size || nblkptr < doi.doi_nblkptr) { - err = dmu_free_long_range(os, drro->drr_object, + err = dmu_free_long_range(ra->os, drro->drr_object, 0, DMU_OBJECT_END); if (err != 0) return (SET_ERROR(EINVAL)); } } - tx = dmu_tx_create(os); + tx = dmu_tx_create(ra->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(os, drro->drr_object, + err = dmu_object_claim(ra->os, drro->drr_object, drro->drr_type, drro->drr_blksz, drro->drr_bonustype, drro->drr_bonuslen, 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) { /* currently allocated, but with different properties */ - err = dmu_object_reclaim(os, drro->drr_object, + err = dmu_object_reclaim(ra->os, drro->drr_object, drro->drr_type, drro->drr_blksz, drro->drr_bonustype, drro->drr_bonuslen, tx); } if (err != 0) { dmu_tx_commit(tx); return (SET_ERROR(EINVAL)); } - dmu_object_set_checksum(os, drro->drr_object, drro->drr_checksumtype, - tx); - dmu_object_set_compress(os, drro->drr_object, drro->drr_compress, tx); + dmu_object_set_checksum(ra->os, drro->drr_object, + drro->drr_checksumtype, tx); + dmu_object_set_compress(ra->os, drro->drr_object, + drro->drr_compress, tx); if (data != NULL) { dmu_buf_t *db; - VERIFY(0 == dmu_bonus_hold(os, drro->drr_object, FTAG, &db)); + VERIFY0(dmu_bonus_hold(ra->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 (ra->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 -restore_freeobjects(struct restorearg *ra, objset_t *os, +restore_freeobjects(struct restorearg *ra, struct drr_freeobjects *drrfo) { uint64_t obj; if (drrfo->drr_firstobj + drrfo->drr_numobjs < drrfo->drr_firstobj) return (SET_ERROR(EINVAL)); for (obj = drrfo->drr_firstobj; obj < drrfo->drr_firstobj + drrfo->drr_numobjs; - (void) dmu_object_next(os, &obj, FALSE, 0)) { + (void) dmu_object_next(ra->os, &obj, FALSE, 0)) { int err; - if (dmu_object_info(os, obj, NULL) != 0) + if (dmu_object_info(ra->os, obj, NULL) != 0) continue; - err = dmu_free_long_object(os, obj); + err = dmu_free_long_object(ra->os, obj); if (err != 0) return (err); } return (0); } static int -restore_write(struct restorearg *ra, objset_t *os, - struct drr_write *drrw) +restore_write(struct restorearg *ra, struct drr_write *drrw, arc_buf_t *abuf) { dmu_tx_t *tx; - void *data; int err; if (drrw->drr_offset + drrw->drr_length < drrw->drr_offset || !DMU_OT_IS_VALID(drrw->drr_type)) return (SET_ERROR(EINVAL)); - if (dmu_object_info(os, drrw->drr_object, NULL) != 0) + if (dmu_object_info(ra->os, drrw->drr_object, NULL) != 0) return (SET_ERROR(EINVAL)); - dmu_buf_t *bonus; - if (dmu_bonus_hold(os, drrw->drr_object, FTAG, &bonus) != 0) - return (SET_ERROR(EINVAL)); + tx = dmu_tx_create(ra->os); - arc_buf_t *abuf = dmu_request_arcbuf(bonus, drrw->drr_length); - - data = restore_read(ra, drrw->drr_length, abuf->b_data); - if (data == NULL) { - dmu_return_arcbuf(abuf); - dmu_buf_rele(bonus, FTAG); - return (ra->err); - } - - tx = dmu_tx_create(os); - dmu_tx_hold_write(tx, drrw->drr_object, drrw->drr_offset, drrw->drr_length); err = dmu_tx_assign(tx, TXG_WAIT); if (err != 0) { - dmu_return_arcbuf(abuf); - dmu_buf_rele(bonus, FTAG); dmu_tx_abort(tx); return (err); } if (ra->byteswap) { dmu_object_byteswap_t byteswap = DMU_OT_BYTESWAP(drrw->drr_type); - dmu_ot_byteswap[byteswap].ob_func(data, drrw->drr_length); + dmu_ot_byteswap[byteswap].ob_func(abuf->b_data, + drrw->drr_length); } + + dmu_buf_t *bonus; + if (dmu_bonus_hold(ra->os, drrw->drr_object, FTAG, &bonus) != 0) + return (SET_ERROR(EINVAL)); dmu_assign_arcbuf(bonus, drrw->drr_offset, abuf, 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 -restore_write_byref(struct restorearg *ra, objset_t *os, - struct drr_write_byref *drrwbr) +restore_write_byref(struct restorearg *ra, 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(ra->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 = os; + ref_os = ra->os; } 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(os); + tx = dmu_tx_create(ra->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(os, drrwbr->drr_object, + dmu_write(ra->os, drrwbr->drr_object, drrwbr->drr_offset, drrwbr->drr_length, dbp->db_data, tx); dmu_buf_rele(dbp, FTAG); dmu_tx_commit(tx); return (0); } static int -restore_write_embedded(struct restorearg *ra, objset_t *os, - struct drr_write_embedded *drrwnp) +restore_write_embedded(struct restorearg *ra, + struct drr_write_embedded *drrwnp, void *data) { dmu_tx_t *tx; int err; - void *data; if (drrwnp->drr_offset + drrwnp->drr_length < drrwnp->drr_offset) return (EINVAL); if (drrwnp->drr_psize > BPE_PAYLOAD_SIZE) return (EINVAL); if (drrwnp->drr_etype >= NUM_BP_EMBEDDED_TYPES) return (EINVAL); if (drrwnp->drr_compression >= ZIO_COMPRESS_FUNCTIONS) return (EINVAL); - data = restore_read(ra, P2ROUNDUP(drrwnp->drr_psize, 8), NULL); - if (data == NULL) - return (ra->err); + tx = dmu_tx_create(ra->os); - tx = dmu_tx_create(os); - dmu_tx_hold_write(tx, drrwnp->drr_object, drrwnp->drr_offset, drrwnp->drr_length); err = dmu_tx_assign(tx, TXG_WAIT); if (err != 0) { dmu_tx_abort(tx); return (err); } - dmu_write_embedded(os, drrwnp->drr_object, + dmu_write_embedded(ra->os, drrwnp->drr_object, drrwnp->drr_offset, data, drrwnp->drr_etype, drrwnp->drr_compression, drrwnp->drr_lsize, drrwnp->drr_psize, ra->byteswap ^ ZFS_HOST_BYTEORDER, tx); dmu_tx_commit(tx); return (0); } static int -restore_spill(struct restorearg *ra, objset_t *os, struct drr_spill *drrs) +restore_spill(struct restorearg *ra, struct drr_spill *drrs, void *data) { dmu_tx_t *tx; - void *data; dmu_buf_t *db, *db_spill; int err; if (drrs->drr_length < SPA_MINBLOCKSIZE || - drrs->drr_length > spa_maxblocksize(dmu_objset_spa(os))) + drrs->drr_length > spa_maxblocksize(dmu_objset_spa(ra->os))) return (SET_ERROR(EINVAL)); - data = restore_read(ra, drrs->drr_length, NULL); - if (data == NULL) - return (ra->err); - - if (dmu_object_info(os, drrs->drr_object, NULL) != 0) + if (dmu_object_info(ra->os, drrs->drr_object, NULL) != 0) return (SET_ERROR(EINVAL)); - VERIFY(0 == dmu_bonus_hold(os, drrs->drr_object, FTAG, &db)); + VERIFY0(dmu_bonus_hold(ra->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(os); + tx = dmu_tx_create(ra->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 -restore_free(struct restorearg *ra, objset_t *os, - struct drr_free *drrf) +restore_free(struct restorearg *ra, 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(os, drrf->drr_object, NULL) != 0) + if (dmu_object_info(ra->os, drrf->drr_object, NULL) != 0) return (SET_ERROR(EINVAL)); - err = dmu_free_long_range(os, drrf->drr_object, + err = dmu_free_long_range(ra->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) { char name[MAXNAMELEN]; dsl_dataset_name(drc->drc_ds, name); dsl_dataset_disown(drc->drc_ds, dmu_recv_tag); (void) dsl_destroy_head(name); } +static void +restore_cksum(struct restorearg *ra, int len, void *buf) +{ + if (ra->byteswap) { + fletcher_4_incremental_byteswap(buf, len, &ra->cksum); + } else { + fletcher_4_incremental_native(buf, len, &ra->cksum); + } +} + /* + * If len != 0, read payload into buf. + * Read next record's header into ra->next_drr. + * Verify checksum of payload and next record. + */ +static int +restore_read_payload_and_next_header(struct restorearg *ra, int len, void *buf) +{ + int err; + + if (len != 0) { + ASSERT3U(len, <=, ra->bufsize); + err = restore_read(ra, len, buf); + if (err != 0) + return (err); + restore_cksum(ra, len, buf); + } + + ra->prev_cksum = ra->cksum; + + err = restore_read(ra, sizeof (*ra->next_drr), ra->next_drr); + if (err != 0) + return (err); + if (ra->next_drr->drr_type == DRR_BEGIN) + 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)); + restore_cksum(ra, + offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), + ra->next_drr); + + zio_cksum_t cksum_orig = ra->next_drr->drr_u.drr_checksum.drr_checksum; + zio_cksum_t *cksump = &ra->next_drr->drr_u.drr_checksum.drr_checksum; + + if (ra->byteswap) + byteswap_record(ra->next_drr); + + if ((!ZIO_CHECKSUM_IS_ZERO(cksump)) && + !ZIO_CHECKSUM_EQUAL(ra->cksum, *cksump)) + return (SET_ERROR(ECKSUM)); + + restore_cksum(ra, sizeof (cksum_orig), &cksum_orig); + + return (0); +} + +static int +restore_process_record(struct restorearg *ra) +{ + int err; + + switch (ra->drr->drr_type) { + case DRR_OBJECT: + { + struct drr_object *drro = &ra->drr->drr_u.drr_object; + err = restore_read_payload_and_next_header(ra, + P2ROUNDUP(drro->drr_bonuslen, 8), ra->buf); + if (err != 0) + return (err); + return (restore_object(ra, drro, ra->buf)); + } + case DRR_FREEOBJECTS: + { + struct drr_freeobjects *drrfo = + &ra->drr->drr_u.drr_freeobjects; + err = restore_read_payload_and_next_header(ra, 0, NULL); + if (err != 0) + return (err); + return (restore_freeobjects(ra, drrfo)); + } + case DRR_WRITE: + { + struct drr_write *drrw = &ra->drr->drr_u.drr_write; + arc_buf_t *abuf = arc_loan_buf(dmu_objset_spa(ra->os), + drrw->drr_length); + + err = restore_read_payload_and_next_header(ra, + drrw->drr_length, abuf->b_data); + if (err != 0) + return (err); + err = restore_write(ra, drrw, abuf); + /* if restore_write() is successful, it consumes the arc_buf */ + if (err != 0) + dmu_return_arcbuf(abuf); + return (err); + } + case DRR_WRITE_BYREF: + { + struct drr_write_byref *drrwbr = + &ra->drr->drr_u.drr_write_byref; + err = restore_read_payload_and_next_header(ra, 0, NULL); + if (err != 0) + return (err); + return (restore_write_byref(ra, drrwbr)); + } + case DRR_WRITE_EMBEDDED: + { + struct drr_write_embedded *drrwe = + &ra->drr->drr_u.drr_write_embedded; + err = restore_read_payload_and_next_header(ra, + P2ROUNDUP(drrwe->drr_psize, 8), ra->buf); + if (err != 0) + return (err); + return (restore_write_embedded(ra, drrwe, ra->buf)); + } + case DRR_FREE: + { + struct drr_free *drrf = &ra->drr->drr_u.drr_free; + err = restore_read_payload_and_next_header(ra, 0, NULL); + if (err != 0) + return (err); + return (restore_free(ra, drrf)); + } + case DRR_END: + { + struct drr_end *drre = &ra->drr->drr_u.drr_end; + if (!ZIO_CHECKSUM_EQUAL(ra->prev_cksum, drre->drr_checksum)) + return (SET_ERROR(EINVAL)); + return (0); + } + case DRR_SPILL: + { + struct drr_spill *drrs = &ra->drr->drr_u.drr_spill; + err = restore_read_payload_and_next_header(ra, + drrs->drr_length, ra->buf); + if (err != 0) + return (err); + return (restore_spill(ra, drrs, ra->buf)); + } + default: + return (SET_ERROR(EINVAL)); + } +} + +/* * NB: callers *must* call dmu_recv_end() if this succeeds. */ int dmu_recv_stream(dmu_recv_cookie_t *drc, struct file *fp, offset_t *voffp, int cleanup_fd, uint64_t *action_handlep) { + int err = 0; struct restorearg ra = { 0 }; - dmu_replay_record_t *drr; - objset_t *os; - zio_cksum_t pcksum; int featureflags; ra.byteswap = drc->drc_byteswap; ra.cksum = drc->drc_cksum; ra.td = curthread; ra.fp = fp; ra.voff = *voffp; ra.bufsize = SPA_MAXBLOCKSIZE; + ra.drr = kmem_alloc(sizeof (*ra.drr), KM_SLEEP); ra.buf = kmem_alloc(ra.bufsize, KM_SLEEP); + ra.next_drr = kmem_alloc(sizeof (*ra.next_drr), KM_SLEEP); /* 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, &os)); + 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) { ra.guid_to_ds_map = kmem_alloc(sizeof (avl_tree_t), KM_SLEEP); avl_create(ra.guid_to_ds_map, guid_compare, sizeof (guid_map_entry_t), offsetof(guid_map_entry_t, avlnode)); - ra.err = zfs_onexit_add_cb(minor, + err = zfs_onexit_add_cb(minor, free_guid_map_onexit, ra.guid_to_ds_map, action_handlep); if (ra.err != 0) goto out; } else { - ra.err = zfs_onexit_cb_data(minor, *action_handlep, + err = zfs_onexit_cb_data(minor, *action_handlep, (void **)&ra.guid_to_ds_map); if (ra.err != 0) goto out; } drc->drc_guid_to_ds_map = ra.guid_to_ds_map; } - /* - * Read records and process them. - */ - pcksum = ra.cksum; - while (ra.err == 0 && - NULL != (drr = restore_read(&ra, sizeof (*drr), NULL))) { + err = restore_read_payload_and_next_header(&ra, 0, NULL); + if (err != 0) + goto out; + for (;;) { + void *tmp; + if (issig(JUSTLOOKING) && issig(FORREAL)) { - ra.err = SET_ERROR(EINTR); - goto out; + err = SET_ERROR(EINTR); + break; } - if (ra.byteswap) - backup_byteswap(drr); + tmp = ra.next_drr; + ra.next_drr = ra.drr; + ra.drr = tmp; - switch (drr->drr_type) { - case DRR_OBJECT: - { - /* - * We need to make a copy of the record header, - * because restore_{object,write} may need to - * restore_read(), which will invalidate drr. - */ - struct drr_object drro = drr->drr_u.drr_object; - ra.err = restore_object(&ra, os, &drro); + /* process ra.drr, read in ra.next_drr */ + err = restore_process_record(&ra); + if (err != 0) break; - } - case DRR_FREEOBJECTS: - { - struct drr_freeobjects drrfo = - drr->drr_u.drr_freeobjects; - ra.err = restore_freeobjects(&ra, os, &drrfo); + if (ra.drr->drr_type == DRR_END) break; - } - case DRR_WRITE: - { - struct drr_write drrw = drr->drr_u.drr_write; - ra.err = restore_write(&ra, os, &drrw); - break; - } - case DRR_WRITE_BYREF: - { - struct drr_write_byref drrwbr = - drr->drr_u.drr_write_byref; - ra.err = restore_write_byref(&ra, os, &drrwbr); - break; - } - case DRR_WRITE_EMBEDDED: - { - struct drr_write_embedded drrwe = - drr->drr_u.drr_write_embedded; - ra.err = restore_write_embedded(&ra, os, &drrwe); - break; - } - case DRR_FREE: - { - struct drr_free drrf = drr->drr_u.drr_free; - ra.err = restore_free(&ra, os, &drrf); - break; - } - case DRR_END: - { - struct drr_end drre = drr->drr_u.drr_end; - /* - * We compare against the *previous* checksum - * value, because the stored checksum is of - * everything before the DRR_END record. - */ - if (!ZIO_CHECKSUM_EQUAL(drre.drr_checksum, pcksum)) - ra.err = SET_ERROR(ECKSUM); - goto out; - } - case DRR_SPILL: - { - struct drr_spill drrs = drr->drr_u.drr_spill; - ra.err = restore_spill(&ra, os, &drrs); - break; - } - default: - ra.err = SET_ERROR(EINVAL); - goto out; - } - pcksum = ra.cksum; } - ASSERT(ra.err != 0); out: if ((featureflags & DMU_BACKUP_FEATURE_DEDUP) && (cleanup_fd != -1)) zfs_onexit_fd_rele(cleanup_fd); - if (ra.err != 0) { + if (err != 0) { /* * destroy what we created, so we don't leave it in the * inconsistent restoring state. */ dmu_recv_cleanup_ds(drc); } + kmem_free(ra.drr, sizeof (*ra.drr)); kmem_free(ra.buf, ra.bufsize); + kmem_free(ra.next_drr, sizeof (*ra.next_drr)); *voffp = ra.voff; - return (ra.err); + 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; 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; } 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) { int error; char name[MAXNAMELEN]; #ifdef _KERNEL /* * We will be destroying the ds; make sure its origin is unmounted if * necessary. */ dsl_dataset_name(drc->drc_ds, name); zfs_destroy_unmount_origin(name); #endif error = dsl_sync_task(drc->drc_tofs, dmu_recv_end_check, dmu_recv_end_sync, drc, dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL); if (error != 0) dmu_recv_cleanup_ds(drc); return (error); } static int dmu_recv_new_end(dmu_recv_cookie_t *drc) { int error; error = dsl_sync_task(drc->drc_tofs, dmu_recv_end_check, dmu_recv_end_sync, drc, dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL); 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); } int dmu_recv_end(dmu_recv_cookie_t *drc, void *owner) { drc->drc_owner = owner; if (drc->drc_newfs) return (dmu_recv_new_end(drc)); else return (dmu_recv_existing_end(drc)); } /* * 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); } Index: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/spa.h =================================================================== --- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/spa.h (revision 286586) +++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/spa.h (revision 286587) @@ -1,899 +1,912 @@ /* * 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) 2011, 2014 by Delphix. All rights reserved. * Copyright 2011 Nexenta Systems, Inc. All rights reserved. * Copyright (c) 2014 Spectra Logic Corporation, All rights reserved. */ #ifndef _SYS_SPA_H #define _SYS_SPA_H #include #include #include #include #include #include #ifdef __cplusplus extern "C" { #endif /* * Forward references that lots of things need. */ typedef struct spa spa_t; typedef struct vdev vdev_t; typedef struct metaslab metaslab_t; typedef struct metaslab_group metaslab_group_t; typedef struct metaslab_class metaslab_class_t; typedef struct zio zio_t; typedef struct zilog zilog_t; typedef struct spa_aux_vdev spa_aux_vdev_t; typedef struct ddt ddt_t; typedef struct ddt_entry ddt_entry_t; struct dsl_pool; struct dsl_dataset; /* * General-purpose 32-bit and 64-bit bitfield encodings. */ #define BF32_DECODE(x, low, len) P2PHASE((x) >> (low), 1U << (len)) #define BF64_DECODE(x, low, len) P2PHASE((x) >> (low), 1ULL << (len)) #define BF32_ENCODE(x, low, len) (P2PHASE((x), 1U << (len)) << (low)) #define BF64_ENCODE(x, low, len) (P2PHASE((x), 1ULL << (len)) << (low)) #define BF32_GET(x, low, len) BF32_DECODE(x, low, len) #define BF64_GET(x, low, len) BF64_DECODE(x, low, len) #define BF32_SET(x, low, len, val) do { \ ASSERT3U(val, <, 1U << (len)); \ ASSERT3U(low + len, <=, 32); \ (x) ^= BF32_ENCODE((x >> low) ^ (val), low, len); \ _NOTE(CONSTCOND) } while (0) #define BF64_SET(x, low, len, val) do { \ ASSERT3U(val, <, 1ULL << (len)); \ ASSERT3U(low + len, <=, 64); \ ((x) ^= BF64_ENCODE((x >> low) ^ (val), low, len)); \ _NOTE(CONSTCOND) } while (0) #define BF32_GET_SB(x, low, len, shift, bias) \ ((BF32_GET(x, low, len) + (bias)) << (shift)) #define BF64_GET_SB(x, low, len, shift, bias) \ ((BF64_GET(x, low, len) + (bias)) << (shift)) #define BF32_SET_SB(x, low, len, shift, bias, val) do { \ ASSERT(IS_P2ALIGNED(val, 1U << shift)); \ ASSERT3S((val) >> (shift), >=, bias); \ BF32_SET(x, low, len, ((val) >> (shift)) - (bias)); \ _NOTE(CONSTCOND) } while (0) #define BF64_SET_SB(x, low, len, shift, bias, val) do { \ ASSERT(IS_P2ALIGNED(val, 1ULL << shift)); \ ASSERT3S((val) >> (shift), >=, bias); \ BF64_SET(x, low, len, ((val) >> (shift)) - (bias)); \ _NOTE(CONSTCOND) } while (0) /* * We currently support block sizes from 512 bytes to 16MB. * The benefits of larger blocks, and thus larger IO, need to be weighed * against the cost of COWing a giant block to modify one byte, and the * large latency of reading or writing a large block. * * Note that although blocks up to 16MB are supported, the recordsize * property can not be set larger than zfs_max_recordsize (default 1MB). * See the comment near zfs_max_recordsize in dsl_dataset.c for details. * * Note that although the LSIZE field of the blkptr_t can store sizes up * to 32MB, the dnode's dn_datablkszsec can only store sizes up to * 32MB - 512 bytes. Therefore, we limit SPA_MAXBLOCKSIZE to 16MB. */ #define SPA_MINBLOCKSHIFT 9 #define SPA_OLD_MAXBLOCKSHIFT 17 #define SPA_MAXBLOCKSHIFT 24 #define SPA_MINBLOCKSIZE (1ULL << SPA_MINBLOCKSHIFT) #define SPA_OLD_MAXBLOCKSIZE (1ULL << SPA_OLD_MAXBLOCKSHIFT) #define SPA_MAXBLOCKSIZE (1ULL << SPA_MAXBLOCKSHIFT) /* * Default maximum supported logical ashift. * * The current 8k allocation block size limit is due to the 8k * aligned/sized operations performed by vdev_probe() on * vdev_label->vl_pad2. Using another "safe region" for these tests * would allow the limit to be raised to 16k, at the expense of * only having 8 available uberblocks in the label area. */ #define SPA_MAXASHIFT 13 /* * Default minimum supported logical ashift. */ #define SPA_MINASHIFT SPA_MINBLOCKSHIFT /* * Size of block to hold the configuration data (a packed nvlist) */ #define SPA_CONFIG_BLOCKSIZE (1ULL << 14) /* * The DVA size encodings for LSIZE and PSIZE support blocks up to 32MB. * The ASIZE encoding should be at least 64 times larger (6 more bits) * to support up to 4-way RAID-Z mirror mode with worst-case gang block * overhead, three DVAs per bp, plus one more bit in case we do anything * else that expands the ASIZE. */ #define SPA_LSIZEBITS 16 /* LSIZE up to 32M (2^16 * 512) */ #define SPA_PSIZEBITS 16 /* PSIZE up to 32M (2^16 * 512) */ #define SPA_ASIZEBITS 24 /* ASIZE up to 64 times larger */ /* * All SPA data is represented by 128-bit data virtual addresses (DVAs). * The members of the dva_t should be considered opaque outside the SPA. */ typedef struct dva { uint64_t dva_word[2]; } dva_t; /* * Each block has a 256-bit checksum -- strong enough for cryptographic hashes. */ typedef struct zio_cksum { uint64_t zc_word[4]; } zio_cksum_t; /* * Each block is described by its DVAs, time of birth, checksum, etc. * The word-by-word, bit-by-bit layout of the blkptr is as follows: * * 64 56 48 40 32 24 16 8 0 * +-------+-------+-------+-------+-------+-------+-------+-------+ * 0 | vdev1 | GRID | ASIZE | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 1 |G| offset1 | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 2 | vdev2 | GRID | ASIZE | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 3 |G| offset2 | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 4 | vdev3 | GRID | ASIZE | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 5 |G| offset3 | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 6 |BDX|lvl| type | cksum |E| comp| PSIZE | LSIZE | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 7 | padding | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 8 | padding | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 9 | physical birth txg | * +-------+-------+-------+-------+-------+-------+-------+-------+ * a | logical birth txg | * +-------+-------+-------+-------+-------+-------+-------+-------+ * b | fill count | * +-------+-------+-------+-------+-------+-------+-------+-------+ * c | checksum[0] | * +-------+-------+-------+-------+-------+-------+-------+-------+ * d | checksum[1] | * +-------+-------+-------+-------+-------+-------+-------+-------+ * e | checksum[2] | * +-------+-------+-------+-------+-------+-------+-------+-------+ * f | checksum[3] | * +-------+-------+-------+-------+-------+-------+-------+-------+ * * Legend: * * vdev virtual device ID * offset offset into virtual device * LSIZE logical size * PSIZE physical size (after compression) * ASIZE allocated size (including RAID-Z parity and gang block headers) * GRID RAID-Z layout information (reserved for future use) * cksum checksum function * comp compression function * G gang block indicator * B byteorder (endianness) * D dedup * X encryption (on version 30, which is not supported) * E blkptr_t contains embedded data (see below) * lvl level of indirection * type DMU object type * phys birth txg of block allocation; zero if same as logical birth txg * log. birth transaction group in which the block was logically born * fill count number of non-zero blocks under this bp * checksum[4] 256-bit checksum of the data this bp describes */ /* * "Embedded" blkptr_t's don't actually point to a block, instead they * have a data payload embedded in the blkptr_t itself. See the comment * in blkptr.c for more details. * * The blkptr_t is laid out as follows: * * 64 56 48 40 32 24 16 8 0 * +-------+-------+-------+-------+-------+-------+-------+-------+ * 0 | payload | * 1 | payload | * 2 | payload | * 3 | payload | * 4 | payload | * 5 | payload | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 6 |BDX|lvl| type | etype |E| comp| PSIZE| LSIZE | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 7 | payload | * 8 | payload | * 9 | payload | * +-------+-------+-------+-------+-------+-------+-------+-------+ * a | logical birth txg | * +-------+-------+-------+-------+-------+-------+-------+-------+ * b | payload | * c | payload | * d | payload | * e | payload | * f | payload | * +-------+-------+-------+-------+-------+-------+-------+-------+ * * Legend: * * payload contains the embedded data * B (byteorder) byteorder (endianness) * D (dedup) padding (set to zero) * X encryption (set to zero; see above) * E (embedded) set to one * lvl indirection level * type DMU object type * etype how to interpret embedded data (BP_EMBEDDED_TYPE_*) * comp compression function of payload * PSIZE size of payload after compression, in bytes * LSIZE logical size of payload, in bytes * note that 25 bits is enough to store the largest * "normal" BP's LSIZE (2^16 * 2^9) in bytes * log. birth transaction group in which the block was logically born * * Note that LSIZE and PSIZE are stored in bytes, whereas for non-embedded * bp's they are stored in units of SPA_MINBLOCKSHIFT. * Generally, the generic BP_GET_*() macros can be used on embedded BP's. * The B, D, X, lvl, type, and comp fields are stored the same as with normal * BP's so the BP_SET_* macros can be used with them. etype, PSIZE, LSIZE must * be set with the BPE_SET_* macros. BP_SET_EMBEDDED() should be called before * other macros, as they assert that they are only used on BP's of the correct * "embedded-ness". */ #define BPE_GET_ETYPE(bp) \ (ASSERT(BP_IS_EMBEDDED(bp)), \ BF64_GET((bp)->blk_prop, 40, 8)) #define BPE_SET_ETYPE(bp, t) do { \ ASSERT(BP_IS_EMBEDDED(bp)); \ BF64_SET((bp)->blk_prop, 40, 8, t); \ _NOTE(CONSTCOND) } while (0) #define BPE_GET_LSIZE(bp) \ (ASSERT(BP_IS_EMBEDDED(bp)), \ BF64_GET_SB((bp)->blk_prop, 0, 25, 0, 1)) #define BPE_SET_LSIZE(bp, x) do { \ ASSERT(BP_IS_EMBEDDED(bp)); \ BF64_SET_SB((bp)->blk_prop, 0, 25, 0, 1, x); \ _NOTE(CONSTCOND) } while (0) #define BPE_GET_PSIZE(bp) \ (ASSERT(BP_IS_EMBEDDED(bp)), \ BF64_GET_SB((bp)->blk_prop, 25, 7, 0, 1)) #define BPE_SET_PSIZE(bp, x) do { \ ASSERT(BP_IS_EMBEDDED(bp)); \ BF64_SET_SB((bp)->blk_prop, 25, 7, 0, 1, x); \ _NOTE(CONSTCOND) } while (0) typedef enum bp_embedded_type { BP_EMBEDDED_TYPE_DATA, BP_EMBEDDED_TYPE_RESERVED, /* Reserved for an unintegrated feature. */ NUM_BP_EMBEDDED_TYPES = BP_EMBEDDED_TYPE_RESERVED } bp_embedded_type_t; #define BPE_NUM_WORDS 14 #define BPE_PAYLOAD_SIZE (BPE_NUM_WORDS * sizeof (uint64_t)) #define BPE_IS_PAYLOADWORD(bp, wp) \ ((wp) != &(bp)->blk_prop && (wp) != &(bp)->blk_birth) #define SPA_BLKPTRSHIFT 7 /* blkptr_t is 128 bytes */ #define SPA_DVAS_PER_BP 3 /* Number of DVAs in a bp */ /* * A block is a hole when it has either 1) never been written to, or * 2) is zero-filled. In both cases, ZFS can return all zeroes for all reads * without physically allocating disk space. Holes are represented in the * blkptr_t structure by zeroed blk_dva. Correct checking for holes is * done through the BP_IS_HOLE macro. For holes, the logical size, level, * DMU object type, and birth times are all also stored for holes that * were written to at some point (i.e. were punched after having been filled). */ typedef struct blkptr { dva_t blk_dva[SPA_DVAS_PER_BP]; /* Data Virtual Addresses */ uint64_t blk_prop; /* size, compression, type, etc */ uint64_t blk_pad[2]; /* Extra space for the future */ uint64_t blk_phys_birth; /* txg when block was allocated */ uint64_t blk_birth; /* transaction group at birth */ uint64_t blk_fill; /* fill count */ zio_cksum_t blk_cksum; /* 256-bit checksum */ } blkptr_t; /* * Macros to get and set fields in a bp or DVA. */ #define DVA_GET_ASIZE(dva) \ BF64_GET_SB((dva)->dva_word[0], 0, SPA_ASIZEBITS, SPA_MINBLOCKSHIFT, 0) #define DVA_SET_ASIZE(dva, x) \ BF64_SET_SB((dva)->dva_word[0], 0, SPA_ASIZEBITS, \ SPA_MINBLOCKSHIFT, 0, x) #define DVA_GET_GRID(dva) BF64_GET((dva)->dva_word[0], 24, 8) #define DVA_SET_GRID(dva, x) BF64_SET((dva)->dva_word[0], 24, 8, x) #define DVA_GET_VDEV(dva) BF64_GET((dva)->dva_word[0], 32, 32) #define DVA_SET_VDEV(dva, x) BF64_SET((dva)->dva_word[0], 32, 32, x) #define DVA_GET_OFFSET(dva) \ BF64_GET_SB((dva)->dva_word[1], 0, 63, SPA_MINBLOCKSHIFT, 0) #define DVA_SET_OFFSET(dva, x) \ BF64_SET_SB((dva)->dva_word[1], 0, 63, SPA_MINBLOCKSHIFT, 0, x) #define DVA_GET_GANG(dva) BF64_GET((dva)->dva_word[1], 63, 1) #define DVA_SET_GANG(dva, x) BF64_SET((dva)->dva_word[1], 63, 1, x) #define BP_GET_LSIZE(bp) \ (BP_IS_EMBEDDED(bp) ? \ (BPE_GET_ETYPE(bp) == BP_EMBEDDED_TYPE_DATA ? BPE_GET_LSIZE(bp) : 0): \ BF64_GET_SB((bp)->blk_prop, 0, SPA_LSIZEBITS, SPA_MINBLOCKSHIFT, 1)) #define BP_SET_LSIZE(bp, x) do { \ ASSERT(!BP_IS_EMBEDDED(bp)); \ BF64_SET_SB((bp)->blk_prop, \ 0, SPA_LSIZEBITS, SPA_MINBLOCKSHIFT, 1, x); \ _NOTE(CONSTCOND) } while (0) #define BP_GET_PSIZE(bp) \ (BP_IS_EMBEDDED(bp) ? 0 : \ BF64_GET_SB((bp)->blk_prop, 16, SPA_PSIZEBITS, SPA_MINBLOCKSHIFT, 1)) #define BP_SET_PSIZE(bp, x) do { \ ASSERT(!BP_IS_EMBEDDED(bp)); \ BF64_SET_SB((bp)->blk_prop, \ 16, SPA_PSIZEBITS, SPA_MINBLOCKSHIFT, 1, x); \ _NOTE(CONSTCOND) } while (0) #define BP_GET_COMPRESS(bp) BF64_GET((bp)->blk_prop, 32, 7) #define BP_SET_COMPRESS(bp, x) BF64_SET((bp)->blk_prop, 32, 7, x) #define BP_IS_EMBEDDED(bp) BF64_GET((bp)->blk_prop, 39, 1) #define BP_SET_EMBEDDED(bp, x) BF64_SET((bp)->blk_prop, 39, 1, x) #define BP_GET_CHECKSUM(bp) \ (BP_IS_EMBEDDED(bp) ? ZIO_CHECKSUM_OFF : \ BF64_GET((bp)->blk_prop, 40, 8)) #define BP_SET_CHECKSUM(bp, x) do { \ ASSERT(!BP_IS_EMBEDDED(bp)); \ BF64_SET((bp)->blk_prop, 40, 8, x); \ _NOTE(CONSTCOND) } while (0) #define BP_GET_TYPE(bp) BF64_GET((bp)->blk_prop, 48, 8) #define BP_SET_TYPE(bp, x) BF64_SET((bp)->blk_prop, 48, 8, x) #define BP_GET_LEVEL(bp) BF64_GET((bp)->blk_prop, 56, 5) #define BP_SET_LEVEL(bp, x) BF64_SET((bp)->blk_prop, 56, 5, x) #define BP_GET_DEDUP(bp) BF64_GET((bp)->blk_prop, 62, 1) #define BP_SET_DEDUP(bp, x) BF64_SET((bp)->blk_prop, 62, 1, x) #define BP_GET_BYTEORDER(bp) BF64_GET((bp)->blk_prop, 63, 1) #define BP_SET_BYTEORDER(bp, x) BF64_SET((bp)->blk_prop, 63, 1, x) #define BP_PHYSICAL_BIRTH(bp) \ (BP_IS_EMBEDDED(bp) ? 0 : \ (bp)->blk_phys_birth ? (bp)->blk_phys_birth : (bp)->blk_birth) #define BP_SET_BIRTH(bp, logical, physical) \ { \ ASSERT(!BP_IS_EMBEDDED(bp)); \ (bp)->blk_birth = (logical); \ (bp)->blk_phys_birth = ((logical) == (physical) ? 0 : (physical)); \ } #define BP_GET_FILL(bp) (BP_IS_EMBEDDED(bp) ? 1 : (bp)->blk_fill) #define BP_GET_ASIZE(bp) \ (BP_IS_EMBEDDED(bp) ? 0 : \ DVA_GET_ASIZE(&(bp)->blk_dva[0]) + \ DVA_GET_ASIZE(&(bp)->blk_dva[1]) + \ DVA_GET_ASIZE(&(bp)->blk_dva[2])) #define BP_GET_UCSIZE(bp) \ ((BP_GET_LEVEL(bp) > 0 || DMU_OT_IS_METADATA(BP_GET_TYPE(bp))) ? \ BP_GET_PSIZE(bp) : BP_GET_LSIZE(bp)) #define BP_GET_NDVAS(bp) \ (BP_IS_EMBEDDED(bp) ? 0 : \ !!DVA_GET_ASIZE(&(bp)->blk_dva[0]) + \ !!DVA_GET_ASIZE(&(bp)->blk_dva[1]) + \ !!DVA_GET_ASIZE(&(bp)->blk_dva[2])) #define BP_COUNT_GANG(bp) \ (BP_IS_EMBEDDED(bp) ? 0 : \ (DVA_GET_GANG(&(bp)->blk_dva[0]) + \ DVA_GET_GANG(&(bp)->blk_dva[1]) + \ DVA_GET_GANG(&(bp)->blk_dva[2]))) #define DVA_EQUAL(dva1, dva2) \ ((dva1)->dva_word[1] == (dva2)->dva_word[1] && \ (dva1)->dva_word[0] == (dva2)->dva_word[0]) #define BP_EQUAL(bp1, bp2) \ (BP_PHYSICAL_BIRTH(bp1) == BP_PHYSICAL_BIRTH(bp2) && \ (bp1)->blk_birth == (bp2)->blk_birth && \ DVA_EQUAL(&(bp1)->blk_dva[0], &(bp2)->blk_dva[0]) && \ DVA_EQUAL(&(bp1)->blk_dva[1], &(bp2)->blk_dva[1]) && \ DVA_EQUAL(&(bp1)->blk_dva[2], &(bp2)->blk_dva[2])) #define ZIO_CHECKSUM_EQUAL(zc1, zc2) \ (0 == (((zc1).zc_word[0] - (zc2).zc_word[0]) | \ ((zc1).zc_word[1] - (zc2).zc_word[1]) | \ ((zc1).zc_word[2] - (zc2).zc_word[2]) | \ ((zc1).zc_word[3] - (zc2).zc_word[3]))) +#define ZIO_CHECKSUM_IS_ZERO(zc) \ + (0 == ((zc)->zc_word[0] | (zc)->zc_word[1] | \ + (zc)->zc_word[2] | (zc)->zc_word[3])) + +#define ZIO_CHECKSUM_BSWAP(zcp) \ +{ \ + (zcp)->zc_word[0] = BSWAP_64((zcp)->zc_word[0]); \ + (zcp)->zc_word[1] = BSWAP_64((zcp)->zc_word[1]); \ + (zcp)->zc_word[2] = BSWAP_64((zcp)->zc_word[2]); \ + (zcp)->zc_word[3] = BSWAP_64((zcp)->zc_word[3]); \ +} + + #define DVA_IS_VALID(dva) (DVA_GET_ASIZE(dva) != 0) #define ZIO_SET_CHECKSUM(zcp, w0, w1, w2, w3) \ { \ (zcp)->zc_word[0] = w0; \ (zcp)->zc_word[1] = w1; \ (zcp)->zc_word[2] = w2; \ (zcp)->zc_word[3] = w3; \ } #define BP_IDENTITY(bp) (ASSERT(!BP_IS_EMBEDDED(bp)), &(bp)->blk_dva[0]) #define BP_IS_GANG(bp) \ (BP_IS_EMBEDDED(bp) ? B_FALSE : DVA_GET_GANG(BP_IDENTITY(bp))) #define DVA_IS_EMPTY(dva) ((dva)->dva_word[0] == 0ULL && \ (dva)->dva_word[1] == 0ULL) #define BP_IS_HOLE(bp) \ (!BP_IS_EMBEDDED(bp) && DVA_IS_EMPTY(BP_IDENTITY(bp))) /* BP_IS_RAIDZ(bp) assumes no block compression */ #define BP_IS_RAIDZ(bp) (DVA_GET_ASIZE(&(bp)->blk_dva[0]) > \ BP_GET_PSIZE(bp)) #define BP_ZERO(bp) \ { \ (bp)->blk_dva[0].dva_word[0] = 0; \ (bp)->blk_dva[0].dva_word[1] = 0; \ (bp)->blk_dva[1].dva_word[0] = 0; \ (bp)->blk_dva[1].dva_word[1] = 0; \ (bp)->blk_dva[2].dva_word[0] = 0; \ (bp)->blk_dva[2].dva_word[1] = 0; \ (bp)->blk_prop = 0; \ (bp)->blk_pad[0] = 0; \ (bp)->blk_pad[1] = 0; \ (bp)->blk_phys_birth = 0; \ (bp)->blk_birth = 0; \ (bp)->blk_fill = 0; \ ZIO_SET_CHECKSUM(&(bp)->blk_cksum, 0, 0, 0, 0); \ } #if BYTE_ORDER == _BIG_ENDIAN #define ZFS_HOST_BYTEORDER (0ULL) #else #define ZFS_HOST_BYTEORDER (1ULL) #endif #define BP_SHOULD_BYTESWAP(bp) (BP_GET_BYTEORDER(bp) != ZFS_HOST_BYTEORDER) #define BP_SPRINTF_LEN 320 /* * This macro allows code sharing between zfs, libzpool, and mdb. * 'func' is either snprintf() or mdb_snprintf(). * 'ws' (whitespace) can be ' ' for single-line format, '\n' for multi-line. */ #define SNPRINTF_BLKPTR(func, ws, buf, size, bp, type, checksum, compress) \ { \ static const char *copyname[] = \ { "zero", "single", "double", "triple" }; \ int len = 0; \ int copies = 0; \ \ if (bp == NULL) { \ len += func(buf + len, size - len, ""); \ } else if (BP_IS_HOLE(bp)) { \ len += func(buf + len, size - len, ""); \ if (bp->blk_birth > 0) { \ len += func(buf + len, size - len, \ " birth=%lluL", \ (u_longlong_t)bp->blk_birth); \ } \ } else if (BP_IS_EMBEDDED(bp)) { \ len = func(buf + len, size - len, \ "EMBEDDED [L%llu %s] et=%u %s " \ "size=%llxL/%llxP birth=%lluL", \ (u_longlong_t)BP_GET_LEVEL(bp), \ type, \ (int)BPE_GET_ETYPE(bp), \ compress, \ (u_longlong_t)BPE_GET_LSIZE(bp), \ (u_longlong_t)BPE_GET_PSIZE(bp), \ (u_longlong_t)bp->blk_birth); \ } else { \ for (int d = 0; d < BP_GET_NDVAS(bp); d++) { \ const dva_t *dva = &bp->blk_dva[d]; \ if (DVA_IS_VALID(dva)) \ copies++; \ len += func(buf + len, size - len, \ "DVA[%d]=<%llu:%llx:%llx>%c", d, \ (u_longlong_t)DVA_GET_VDEV(dva), \ (u_longlong_t)DVA_GET_OFFSET(dva), \ (u_longlong_t)DVA_GET_ASIZE(dva), \ ws); \ } \ if (BP_IS_GANG(bp) && \ DVA_GET_ASIZE(&bp->blk_dva[2]) <= \ DVA_GET_ASIZE(&bp->blk_dva[1]) / 2) \ copies--; \ len += func(buf + len, size - len, \ "[L%llu %s] %s %s %s %s %s %s%c" \ "size=%llxL/%llxP birth=%lluL/%lluP fill=%llu%c" \ "cksum=%llx:%llx:%llx:%llx", \ (u_longlong_t)BP_GET_LEVEL(bp), \ type, \ checksum, \ compress, \ BP_GET_BYTEORDER(bp) == 0 ? "BE" : "LE", \ BP_IS_GANG(bp) ? "gang" : "contiguous", \ BP_GET_DEDUP(bp) ? "dedup" : "unique", \ copyname[copies], \ ws, \ (u_longlong_t)BP_GET_LSIZE(bp), \ (u_longlong_t)BP_GET_PSIZE(bp), \ (u_longlong_t)bp->blk_birth, \ (u_longlong_t)BP_PHYSICAL_BIRTH(bp), \ (u_longlong_t)BP_GET_FILL(bp), \ ws, \ (u_longlong_t)bp->blk_cksum.zc_word[0], \ (u_longlong_t)bp->blk_cksum.zc_word[1], \ (u_longlong_t)bp->blk_cksum.zc_word[2], \ (u_longlong_t)bp->blk_cksum.zc_word[3]); \ } \ ASSERT(len < size); \ } #include #define BP_GET_BUFC_TYPE(bp) \ (((BP_GET_LEVEL(bp) > 0) || (DMU_OT_IS_METADATA(BP_GET_TYPE(bp)))) ? \ ARC_BUFC_METADATA : ARC_BUFC_DATA) typedef enum spa_import_type { SPA_IMPORT_EXISTING, SPA_IMPORT_ASSEMBLE } spa_import_type_t; /* state manipulation functions */ extern int spa_open(const char *pool, spa_t **, void *tag); extern int spa_open_rewind(const char *pool, spa_t **, void *tag, nvlist_t *policy, nvlist_t **config); extern int spa_get_stats(const char *pool, nvlist_t **config, char *altroot, size_t buflen); extern int spa_create(const char *pool, nvlist_t *config, nvlist_t *props, nvlist_t *zplprops); #ifdef illumos extern int spa_import_rootpool(char *devpath, char *devid); #else extern int spa_import_rootpool(const char *name); #endif extern int spa_import(const char *pool, nvlist_t *config, nvlist_t *props, uint64_t flags); extern nvlist_t *spa_tryimport(nvlist_t *tryconfig); extern int spa_destroy(char *pool); extern int spa_export(char *pool, nvlist_t **oldconfig, boolean_t force, boolean_t hardforce); extern int spa_reset(char *pool); extern void spa_async_request(spa_t *spa, int flag); extern void spa_async_unrequest(spa_t *spa, int flag); extern void spa_async_suspend(spa_t *spa); extern void spa_async_resume(spa_t *spa); extern spa_t *spa_inject_addref(char *pool); extern void spa_inject_delref(spa_t *spa); extern void spa_scan_stat_init(spa_t *spa); extern int spa_scan_get_stats(spa_t *spa, pool_scan_stat_t *ps); #define SPA_ASYNC_CONFIG_UPDATE 0x01 #define SPA_ASYNC_REMOVE 0x02 #define SPA_ASYNC_PROBE 0x04 #define SPA_ASYNC_RESILVER_DONE 0x08 #define SPA_ASYNC_RESILVER 0x10 #define SPA_ASYNC_AUTOEXPAND 0x20 #define SPA_ASYNC_REMOVE_DONE 0x40 #define SPA_ASYNC_REMOVE_STOP 0x80 /* * Controls the behavior of spa_vdev_remove(). */ #define SPA_REMOVE_UNSPARE 0x01 #define SPA_REMOVE_DONE 0x02 /* device manipulation */ extern int spa_vdev_add(spa_t *spa, nvlist_t *nvroot); extern int spa_vdev_attach(spa_t *spa, uint64_t guid, nvlist_t *nvroot, int replacing); extern int spa_vdev_detach(spa_t *spa, uint64_t guid, uint64_t pguid, int replace_done); extern int spa_vdev_remove(spa_t *spa, uint64_t guid, boolean_t unspare); extern boolean_t spa_vdev_remove_active(spa_t *spa); extern int spa_vdev_setpath(spa_t *spa, uint64_t guid, const char *newpath); extern int spa_vdev_setfru(spa_t *spa, uint64_t guid, const char *newfru); extern int spa_vdev_split_mirror(spa_t *spa, char *newname, nvlist_t *config, nvlist_t *props, boolean_t exp); /* spare state (which is global across all pools) */ extern void spa_spare_add(vdev_t *vd); extern void spa_spare_remove(vdev_t *vd); extern boolean_t spa_spare_exists(uint64_t guid, uint64_t *pool, int *refcnt); extern void spa_spare_activate(vdev_t *vd); /* L2ARC state (which is global across all pools) */ extern void spa_l2cache_add(vdev_t *vd); extern void spa_l2cache_remove(vdev_t *vd); extern boolean_t spa_l2cache_exists(uint64_t guid, uint64_t *pool); extern void spa_l2cache_activate(vdev_t *vd); extern void spa_l2cache_drop(spa_t *spa); /* scanning */ extern int spa_scan(spa_t *spa, pool_scan_func_t func); extern int spa_scan_stop(spa_t *spa); /* spa syncing */ extern void spa_sync(spa_t *spa, uint64_t txg); /* only for DMU use */ extern void spa_sync_allpools(void); /* spa namespace global mutex */ extern kmutex_t spa_namespace_lock; /* * SPA configuration functions in spa_config.c */ #define SPA_CONFIG_UPDATE_POOL 0 #define SPA_CONFIG_UPDATE_VDEVS 1 extern void spa_config_sync(spa_t *, boolean_t, boolean_t); extern void spa_config_load(void); extern nvlist_t *spa_all_configs(uint64_t *); extern void spa_config_set(spa_t *spa, nvlist_t *config); extern nvlist_t *spa_config_generate(spa_t *spa, vdev_t *vd, uint64_t txg, int getstats); extern void spa_config_update(spa_t *spa, int what); /* * Miscellaneous SPA routines in spa_misc.c */ /* Namespace manipulation */ extern spa_t *spa_lookup(const char *name); extern spa_t *spa_add(const char *name, nvlist_t *config, const char *altroot); extern void spa_remove(spa_t *spa); extern spa_t *spa_next(spa_t *prev); /* Refcount functions */ extern void spa_open_ref(spa_t *spa, void *tag); extern void spa_close(spa_t *spa, void *tag); extern void spa_async_close(spa_t *spa, void *tag); extern boolean_t spa_refcount_zero(spa_t *spa); #define SCL_NONE 0x00 #define SCL_CONFIG 0x01 #define SCL_STATE 0x02 #define SCL_L2ARC 0x04 /* hack until L2ARC 2.0 */ #define SCL_ALLOC 0x08 #define SCL_ZIO 0x10 #define SCL_FREE 0x20 #define SCL_VDEV 0x40 #define SCL_LOCKS 7 #define SCL_ALL ((1 << SCL_LOCKS) - 1) #define SCL_STATE_ALL (SCL_STATE | SCL_L2ARC | SCL_ZIO) /* Pool configuration locks */ extern int spa_config_tryenter(spa_t *spa, int locks, void *tag, krw_t rw); extern void spa_config_enter(spa_t *spa, int locks, void *tag, krw_t rw); extern void spa_config_exit(spa_t *spa, int locks, void *tag); extern int spa_config_held(spa_t *spa, int locks, krw_t rw); /* Pool vdev add/remove lock */ extern uint64_t spa_vdev_enter(spa_t *spa); extern uint64_t spa_vdev_config_enter(spa_t *spa); extern void spa_vdev_config_exit(spa_t *spa, vdev_t *vd, uint64_t txg, int error, char *tag); extern int spa_vdev_exit(spa_t *spa, vdev_t *vd, uint64_t txg, int error); /* Pool vdev state change lock */ extern void spa_vdev_state_enter(spa_t *spa, int oplock); extern int spa_vdev_state_exit(spa_t *spa, vdev_t *vd, int error); /* Log state */ typedef enum spa_log_state { SPA_LOG_UNKNOWN = 0, /* unknown log state */ SPA_LOG_MISSING, /* missing log(s) */ SPA_LOG_CLEAR, /* clear the log(s) */ SPA_LOG_GOOD, /* log(s) are good */ } spa_log_state_t; extern spa_log_state_t spa_get_log_state(spa_t *spa); extern void spa_set_log_state(spa_t *spa, spa_log_state_t state); extern int spa_offline_log(spa_t *spa); /* Log claim callback */ extern void spa_claim_notify(zio_t *zio); /* Accessor functions */ extern boolean_t spa_shutting_down(spa_t *spa); extern struct dsl_pool *spa_get_dsl(spa_t *spa); extern boolean_t spa_is_initializing(spa_t *spa); extern blkptr_t *spa_get_rootblkptr(spa_t *spa); extern void spa_set_rootblkptr(spa_t *spa, const blkptr_t *bp); extern void spa_altroot(spa_t *, char *, size_t); extern int spa_sync_pass(spa_t *spa); extern char *spa_name(spa_t *spa); extern uint64_t spa_guid(spa_t *spa); extern uint64_t spa_load_guid(spa_t *spa); extern uint64_t spa_last_synced_txg(spa_t *spa); extern uint64_t spa_first_txg(spa_t *spa); extern uint64_t spa_syncing_txg(spa_t *spa); extern uint64_t spa_version(spa_t *spa); extern pool_state_t spa_state(spa_t *spa); extern spa_load_state_t spa_load_state(spa_t *spa); extern uint64_t spa_freeze_txg(spa_t *spa); extern uint64_t spa_get_asize(spa_t *spa, uint64_t lsize); extern uint64_t spa_get_dspace(spa_t *spa); extern uint64_t spa_get_slop_space(spa_t *spa); extern void spa_update_dspace(spa_t *spa); extern uint64_t spa_version(spa_t *spa); extern boolean_t spa_deflate(spa_t *spa); extern metaslab_class_t *spa_normal_class(spa_t *spa); extern metaslab_class_t *spa_log_class(spa_t *spa); extern void spa_evicting_os_register(spa_t *, objset_t *os); extern void spa_evicting_os_deregister(spa_t *, objset_t *os); extern void spa_evicting_os_wait(spa_t *spa); extern int spa_max_replication(spa_t *spa); extern int spa_prev_software_version(spa_t *spa); extern int spa_busy(void); extern uint8_t spa_get_failmode(spa_t *spa); extern boolean_t spa_suspended(spa_t *spa); extern uint64_t spa_bootfs(spa_t *spa); extern uint64_t spa_delegation(spa_t *spa); extern objset_t *spa_meta_objset(spa_t *spa); extern uint64_t spa_deadman_synctime(spa_t *spa); /* Miscellaneous support routines */ extern void spa_activate_mos_feature(spa_t *spa, const char *feature, dmu_tx_t *tx); extern void spa_deactivate_mos_feature(spa_t *spa, const char *feature); extern int spa_rename(const char *oldname, const char *newname); extern spa_t *spa_by_guid(uint64_t pool_guid, uint64_t device_guid); extern boolean_t spa_guid_exists(uint64_t pool_guid, uint64_t device_guid); extern char *spa_strdup(const char *); extern void spa_strfree(char *); extern uint64_t spa_get_random(uint64_t range); extern uint64_t spa_generate_guid(spa_t *spa); extern void snprintf_blkptr(char *buf, size_t buflen, const blkptr_t *bp); extern void spa_freeze(spa_t *spa); extern int spa_change_guid(spa_t *spa); extern void spa_upgrade(spa_t *spa, uint64_t version); extern void spa_evict_all(void); extern vdev_t *spa_lookup_by_guid(spa_t *spa, uint64_t guid, boolean_t l2cache); extern boolean_t spa_has_spare(spa_t *, uint64_t guid); extern uint64_t dva_get_dsize_sync(spa_t *spa, const dva_t *dva); extern uint64_t bp_get_dsize_sync(spa_t *spa, const blkptr_t *bp); extern uint64_t bp_get_dsize(spa_t *spa, const blkptr_t *bp); extern boolean_t spa_has_slogs(spa_t *spa); extern boolean_t spa_is_root(spa_t *spa); extern boolean_t spa_writeable(spa_t *spa); extern boolean_t spa_has_pending_synctask(spa_t *spa); extern int spa_maxblocksize(spa_t *spa); extern void zfs_blkptr_verify(spa_t *spa, const blkptr_t *bp); extern int spa_mode(spa_t *spa); extern uint64_t zfs_strtonum(const char *str, char **nptr); #define strtonum(str, nptr) zfs_strtonum((str), (nptr)) extern char *spa_his_ievent_table[]; extern void spa_history_create_obj(spa_t *spa, dmu_tx_t *tx); extern int spa_history_get(spa_t *spa, uint64_t *offset, uint64_t *len_read, char *his_buf); extern int spa_history_log(spa_t *spa, const char *his_buf); extern int spa_history_log_nvl(spa_t *spa, nvlist_t *nvl); extern void spa_history_log_version(spa_t *spa, const char *operation); extern void spa_history_log_internal(spa_t *spa, const char *operation, dmu_tx_t *tx, const char *fmt, ...); extern void spa_history_log_internal_ds(struct dsl_dataset *ds, const char *op, dmu_tx_t *tx, const char *fmt, ...); extern void spa_history_log_internal_dd(dsl_dir_t *dd, const char *operation, dmu_tx_t *tx, const char *fmt, ...); /* error handling */ struct zbookmark_phys; extern void spa_log_error(spa_t *spa, zio_t *zio); extern void zfs_ereport_post(const char *cls, spa_t *spa, vdev_t *vd, zio_t *zio, uint64_t stateoroffset, uint64_t length); extern void zfs_post_remove(spa_t *spa, vdev_t *vd); extern void zfs_post_state_change(spa_t *spa, vdev_t *vd); extern void zfs_post_autoreplace(spa_t *spa, vdev_t *vd); extern uint64_t spa_get_errlog_size(spa_t *spa); extern int spa_get_errlog(spa_t *spa, void *uaddr, size_t *count); extern void spa_errlog_rotate(spa_t *spa); extern void spa_errlog_drain(spa_t *spa); extern void spa_errlog_sync(spa_t *spa, uint64_t txg); extern void spa_get_errlists(spa_t *spa, avl_tree_t *last, avl_tree_t *scrub); /* vdev cache */ extern void vdev_cache_stat_init(void); extern void vdev_cache_stat_fini(void); /* Initialization and termination */ extern void spa_init(int flags); extern void spa_fini(void); extern void spa_boot_init(); /* properties */ extern int spa_prop_set(spa_t *spa, nvlist_t *nvp); extern int spa_prop_get(spa_t *spa, nvlist_t **nvp); extern void spa_prop_clear_bootfs(spa_t *spa, uint64_t obj, dmu_tx_t *tx); extern void spa_configfile_set(spa_t *, nvlist_t *, boolean_t); /* asynchronous event notification */ extern void spa_event_notify(spa_t *spa, vdev_t *vdev, const char *name); #ifdef ZFS_DEBUG #define dprintf_bp(bp, fmt, ...) do { \ if (zfs_flags & ZFS_DEBUG_DPRINTF) { \ char *__blkbuf = kmem_alloc(BP_SPRINTF_LEN, KM_SLEEP); \ snprintf_blkptr(__blkbuf, BP_SPRINTF_LEN, (bp)); \ dprintf(fmt " %s\n", __VA_ARGS__, __blkbuf); \ kmem_free(__blkbuf, BP_SPRINTF_LEN); \ } \ _NOTE(CONSTCOND) } while (0) #else #define dprintf_bp(bp, fmt, ...) #endif extern boolean_t spa_debug_enabled(spa_t *spa); #define spa_dbgmsg(spa, ...) \ { \ if (spa_debug_enabled(spa)) \ zfs_dbgmsg(__VA_ARGS__); \ } extern int spa_mode_global; /* mode, e.g. FREAD | FWRITE */ #ifdef __cplusplus } #endif #endif /* _SYS_SPA_H */ Index: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zfs_ioctl.h =================================================================== --- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zfs_ioctl.h (revision 286586) +++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zfs_ioctl.h (revision 286587) @@ -1,413 +1,429 @@ /* * 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) 2013 by Delphix. All rights reserved. + * Copyright (c) 2012, 2014 by Delphix. All rights reserved. */ #ifndef _SYS_ZFS_IOCTL_H #define _SYS_ZFS_IOCTL_H #include #include #include #include #include #include #ifdef _KERNEL #include #endif /* _KERNEL */ #ifdef __cplusplus extern "C" { #endif /* * The structures in this file are passed between userland and the * kernel. Userland may be running a 32-bit process, while the kernel * is 64-bit. Therefore, these structures need to compile the same in * 32-bit and 64-bit. This means not using type "long", and adding * explicit padding so that the 32-bit structure will not be packed more * tightly than the 64-bit structure (which requires 64-bit alignment). */ /* * Property values for snapdir */ #define ZFS_SNAPDIR_HIDDEN 0 #define ZFS_SNAPDIR_VISIBLE 1 /* * Field manipulation macros for the drr_versioninfo field of the * send stream header. */ /* * Header types for zfs send streams. */ typedef enum drr_headertype { DMU_SUBSTREAM = 0x1, DMU_COMPOUNDSTREAM = 0x2 } drr_headertype_t; #define DMU_GET_STREAM_HDRTYPE(vi) BF64_GET((vi), 0, 2) #define DMU_SET_STREAM_HDRTYPE(vi, x) BF64_SET((vi), 0, 2, x) #define DMU_GET_FEATUREFLAGS(vi) BF64_GET((vi), 2, 30) #define DMU_SET_FEATUREFLAGS(vi, x) BF64_SET((vi), 2, 30, x) /* * Feature flags for zfs send streams (flags in drr_versioninfo) */ #define DMU_BACKUP_FEATURE_DEDUP (1<<0) #define DMU_BACKUP_FEATURE_DEDUPPROPS (1<<1) #define DMU_BACKUP_FEATURE_SA_SPILL (1<<2) /* flags #3 - #15 are reserved for incompatible closed-source implementations */ #define DMU_BACKUP_FEATURE_EMBED_DATA (1<<16) #define DMU_BACKUP_FEATURE_EMBED_DATA_LZ4 (1<<17) /* flag #18 is reserved for a Delphix feature */ #define DMU_BACKUP_FEATURE_LARGE_BLOCKS (1<<19) /* * Mask of all supported backup features */ #define DMU_BACKUP_FEATURE_MASK (DMU_BACKUP_FEATURE_DEDUP | \ DMU_BACKUP_FEATURE_DEDUPPROPS | DMU_BACKUP_FEATURE_SA_SPILL | \ DMU_BACKUP_FEATURE_EMBED_DATA | DMU_BACKUP_FEATURE_EMBED_DATA_LZ4 | \ DMU_BACKUP_FEATURE_LARGE_BLOCKS) /* Are all features in the given flag word currently supported? */ #define DMU_STREAM_SUPPORTED(x) (!((x) & ~DMU_BACKUP_FEATURE_MASK)) /* * The drr_versioninfo field of the dmu_replay_record has the * following layout: * * 64 56 48 40 32 24 16 8 0 * +-------+-------+-------+-------+-------+-------+-------+-------+ * | reserved | feature-flags |C|S| * +-------+-------+-------+-------+-------+-------+-------+-------+ * * The low order two bits indicate the header type: SUBSTREAM (0x1) * or COMPOUNDSTREAM (0x2). Using two bits for this is historical: * this field used to be a version number, where the two version types * were 1 and 2. Using two bits for this allows earlier versions of * the code to be able to recognize send streams that don't use any * of the features indicated by feature flags. */ #define DMU_BACKUP_MAGIC 0x2F5bacbacULL #define DRR_FLAG_CLONE (1<<0) #define DRR_FLAG_CI_DATA (1<<1) /* * flags in the drr_checksumflags field in the DRR_WRITE and * DRR_WRITE_BYREF blocks */ #define DRR_CHECKSUM_DEDUP (1<<0) #define DRR_IS_DEDUP_CAPABLE(flags) ((flags) & DRR_CHECKSUM_DEDUP) /* * zfs ioctl command structure */ struct drr_begin { uint64_t drr_magic; uint64_t drr_versioninfo; /* was drr_version */ uint64_t drr_creation_time; dmu_objset_type_t drr_type; uint32_t drr_flags; uint64_t drr_toguid; uint64_t drr_fromguid; char drr_toname[MAXNAMELEN]; }; struct drr_end { zio_cksum_t drr_checksum; uint64_t drr_toguid; }; struct drr_object { uint64_t drr_object; dmu_object_type_t drr_type; dmu_object_type_t drr_bonustype; uint32_t drr_blksz; uint32_t drr_bonuslen; uint8_t drr_checksumtype; uint8_t drr_compress; uint8_t drr_pad[6]; uint64_t drr_toguid; /* bonus content follows */ }; struct drr_freeobjects { uint64_t drr_firstobj; uint64_t drr_numobjs; uint64_t drr_toguid; }; struct drr_write { uint64_t drr_object; dmu_object_type_t drr_type; uint32_t drr_pad; uint64_t drr_offset; uint64_t drr_length; uint64_t drr_toguid; uint8_t drr_checksumtype; uint8_t drr_checksumflags; uint8_t drr_pad2[6]; ddt_key_t drr_key; /* deduplication key */ /* content follows */ }; struct drr_free { uint64_t drr_object; uint64_t drr_offset; uint64_t drr_length; uint64_t drr_toguid; }; struct drr_write_byref { /* where to put the data */ uint64_t drr_object; uint64_t drr_offset; uint64_t drr_length; uint64_t drr_toguid; /* where to find the prior copy of the data */ uint64_t drr_refguid; uint64_t drr_refobject; uint64_t drr_refoffset; /* properties of the data */ uint8_t drr_checksumtype; uint8_t drr_checksumflags; uint8_t drr_pad2[6]; ddt_key_t drr_key; /* deduplication key */ }; struct drr_spill { uint64_t drr_object; uint64_t drr_length; uint64_t drr_toguid; uint64_t drr_pad[4]; /* needed for crypto */ /* spill data follows */ }; typedef struct dmu_replay_record { enum { DRR_BEGIN, DRR_OBJECT, DRR_FREEOBJECTS, DRR_WRITE, DRR_FREE, DRR_END, DRR_WRITE_BYREF, DRR_SPILL, DRR_WRITE_EMBEDDED, DRR_NUMTYPES } drr_type; uint32_t drr_payloadlen; union { struct drr_begin drr_begin; struct drr_end drr_end; struct drr_object drr_object; struct drr_freeobjects drr_freeobjects; struct drr_write drr_write; struct drr_free drr_free; struct drr_write_byref drr_write_byref; struct drr_spill drr_spill; struct drr_write_embedded { uint64_t drr_object; uint64_t drr_offset; /* logical length, should equal blocksize */ uint64_t drr_length; uint64_t drr_toguid; uint8_t drr_compression; uint8_t drr_etype; uint8_t drr_pad[6]; uint32_t drr_lsize; /* uncompressed size of payload */ uint32_t drr_psize; /* compr. (real) size of payload */ /* (possibly compressed) content follows */ } drr_write_embedded; + + /* + * Nore: drr_checksum is overlaid with all record types + * except DRR_BEGIN. Therefore its (non-pad) members + * must not overlap with members from the other structs. + * We accomplish this by putting its members at the very + * end of the struct. + */ + struct drr_checksum { + uint64_t drr_pad[34]; + /* + * fletcher-4 checksum of everything preceding the + * checksum. + */ + zio_cksum_t drr_checksum; + } drr_checksum; } drr_u; } dmu_replay_record_t; /* diff record range types */ typedef enum diff_type { DDR_NONE = 0x1, DDR_INUSE = 0x2, DDR_FREE = 0x4 } diff_type_t; /* * The diff reports back ranges of free or in-use objects. */ typedef struct dmu_diff_record { uint64_t ddr_type; uint64_t ddr_first; uint64_t ddr_last; } dmu_diff_record_t; typedef struct zinject_record { uint64_t zi_objset; uint64_t zi_object; uint64_t zi_start; uint64_t zi_end; uint64_t zi_guid; uint32_t zi_level; uint32_t zi_error; uint64_t zi_type; uint32_t zi_freq; uint32_t zi_failfast; char zi_func[MAXNAMELEN]; uint32_t zi_iotype; int32_t zi_duration; uint64_t zi_timer; uint32_t zi_cmd; uint32_t zi_pad; } zinject_record_t; #define ZINJECT_NULL 0x1 #define ZINJECT_FLUSH_ARC 0x2 #define ZINJECT_UNLOAD_SPA 0x4 typedef enum zinject_type { ZINJECT_UNINITIALIZED, ZINJECT_DATA_FAULT, ZINJECT_DEVICE_FAULT, ZINJECT_LABEL_FAULT, ZINJECT_IGNORED_WRITES, ZINJECT_PANIC, ZINJECT_DELAY_IO, } zinject_type_t; typedef struct zfs_share { uint64_t z_exportdata; uint64_t z_sharedata; uint64_t z_sharetype; /* 0 = share, 1 = unshare */ uint64_t z_sharemax; /* max length of share string */ } zfs_share_t; /* * ZFS file systems may behave the usual, POSIX-compliant way, where * name lookups are case-sensitive. They may also be set up so that * all the name lookups are case-insensitive, or so that only some * lookups, the ones that set an FIGNORECASE flag, are case-insensitive. */ typedef enum zfs_case { ZFS_CASE_SENSITIVE, ZFS_CASE_INSENSITIVE, ZFS_CASE_MIXED } zfs_case_t; typedef struct zfs_cmd { char zc_name[MAXPATHLEN]; /* name of pool or dataset */ uint64_t zc_nvlist_src; /* really (char *) */ uint64_t zc_nvlist_src_size; uint64_t zc_nvlist_dst; /* really (char *) */ uint64_t zc_nvlist_dst_size; boolean_t zc_nvlist_dst_filled; /* put an nvlist in dst? */ int zc_pad2; /* * The following members are for legacy ioctls which haven't been * converted to the new method. */ uint64_t zc_history; /* really (char *) */ char zc_value[MAXPATHLEN * 2]; char zc_string[MAXNAMELEN]; uint64_t zc_guid; uint64_t zc_nvlist_conf; /* really (char *) */ uint64_t zc_nvlist_conf_size; uint64_t zc_cookie; uint64_t zc_objset_type; uint64_t zc_perm_action; uint64_t zc_history_len; uint64_t zc_history_offset; uint64_t zc_obj; uint64_t zc_iflags; /* internal to zfs(7fs) */ zfs_share_t zc_share; uint64_t zc_jailid; dmu_objset_stats_t zc_objset_stats; struct drr_begin zc_begin_record; zinject_record_t zc_inject_record; uint32_t zc_defer_destroy; uint32_t zc_flags; uint64_t zc_action_handle; int zc_cleanup_fd; uint8_t zc_simple; uint8_t zc_pad[3]; /* alignment */ uint64_t zc_sendobj; uint64_t zc_fromobj; uint64_t zc_createtxg; zfs_stat_t zc_stat; } zfs_cmd_t; typedef struct zfs_useracct { char zu_domain[256]; uid_t zu_rid; uint32_t zu_pad; uint64_t zu_space; } zfs_useracct_t; #define ZFSDEV_MAX_MINOR (1 << 16) #define ZFS_MIN_MINOR (ZFSDEV_MAX_MINOR + 1) #define ZPOOL_EXPORT_AFTER_SPLIT 0x1 #ifdef _KERNEL typedef struct zfs_creat { nvlist_t *zct_zplprops; nvlist_t *zct_props; } zfs_creat_t; extern int zfs_secpolicy_snapshot_perms(const char *name, cred_t *cr); extern int zfs_secpolicy_rename_perms(const char *from, const char *to, cred_t *cr); extern int zfs_secpolicy_destroy_perms(const char *name, cred_t *cr); extern int zfs_busy(void); extern int zfs_unmount_snap(const char *); extern void zfs_destroy_unmount_origin(const char *); /* * ZFS minor numbers can refer to either a control device instance or * a zvol. Depending on the value of zss_type, zss_data points to either * a zvol_state_t or a zfs_onexit_t. */ enum zfs_soft_state_type { ZSST_ZVOL, ZSST_CTLDEV }; typedef struct zfs_soft_state { enum zfs_soft_state_type zss_type; void *zss_data; } zfs_soft_state_t; extern void *zfsdev_get_soft_state(minor_t minor, enum zfs_soft_state_type which); extern minor_t zfsdev_minor_alloc(void); extern void *zfsdev_state; #endif /* _KERNEL */ #ifdef __cplusplus } #endif #endif /* _SYS_ZFS_IOCTL_H */ Index: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio_checksum.h =================================================================== --- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio_checksum.h (revision 286586) +++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio_checksum.h (revision 286587) @@ -1,75 +1,76 @@ /* * 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) 2014 by Delphix. All rights reserved. */ #ifndef _SYS_ZIO_CHECKSUM_H #define _SYS_ZIO_CHECKSUM_H #include #ifdef __cplusplus extern "C" { #endif /* * Signature for checksum functions. */ -typedef void zio_checksum_t(const void *data, uint64_t size, zio_cksum_t *zcp); +typedef void zio_checksum_func_t(const void *, uint64_t, zio_cksum_t *); /* * Information about each checksum function. */ typedef struct zio_checksum_info { - zio_checksum_t *ci_func[2]; /* checksum function for each byteorder */ + zio_checksum_func_t *ci_func[2]; /* checksum function per byteorder */ int ci_correctable; /* number of correctable bits */ int ci_eck; /* uses zio embedded checksum? */ int ci_dedup; /* strong enough for dedup? */ char *ci_name; /* descriptive name */ } zio_checksum_info_t; typedef struct zio_bad_cksum { zio_cksum_t zbc_expected; zio_cksum_t zbc_actual; const char *zbc_checksum_name; uint8_t zbc_byteswapped; uint8_t zbc_injected; uint8_t zbc_has_cksum; /* expected/actual valid */ } zio_bad_cksum_t; extern zio_checksum_info_t zio_checksum_table[ZIO_CHECKSUM_FUNCTIONS]; /* * Checksum routines. */ -extern zio_checksum_t zio_checksum_SHA256; +extern zio_checksum_func_t zio_checksum_SHA256; extern void zio_checksum_compute(zio_t *zio, enum zio_checksum checksum, void *data, uint64_t size); extern int zio_checksum_error(zio_t *zio, zio_bad_cksum_t *out); extern enum zio_checksum spa_dedup_checksum(spa_t *spa); #ifdef __cplusplus } #endif #endif /* _SYS_ZIO_CHECKSUM_H */