diff --git a/include/sys/spa.h b/include/sys/spa.h index b96a9ef1d42f..460ea2bfee4e 100644 --- a/include/sys/spa.h +++ b/include/sys/spa.h @@ -1,1238 +1,1239 @@ /* * 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 https://opensource.org/licenses/CDDL-1.0. * 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, 2021 by Delphix. All rights reserved. * Copyright 2011 Nexenta Systems, Inc. All rights reserved. * Copyright (c) 2014 Spectra Logic Corporation, All rights reserved. * Copyright 2013 Saso Kiselkov. All rights reserved. * Copyright (c) 2014 Integros [integros.com] * Copyright 2017 Joyent, Inc. * Copyright (c) 2017, Intel Corporation. * Copyright (c) 2019, Allan Jude * Copyright (c) 2019, Klara Inc. * Copyright (c) 2019, Datto Inc. */ #ifndef _SYS_SPA_H #define _SYS_SPA_H #include #include #include #include #include #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; typedef struct zbookmark_phys zbookmark_phys_t; typedef struct zbookmark_err_phys zbookmark_err_phys_t; struct bpobj; struct bplist; struct dsl_pool; struct dsl_dataset; struct dsl_crypto_params; /* * Alignment Shift (ashift) is an immutable, internal top-level vdev property * which can only be set at vdev creation time. Physical writes are always done * according to it, which makes 2^ashift the smallest possible IO on a vdev. * * We currently allow values ranging from 512 bytes (2^9 = 512) to 64 KiB * (2^16 = 65,536). */ #define ASHIFT_MIN 9 #define ASHIFT_MAX 16 /* * 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 */ #define SPA_COMPRESSBITS 7 #define SPA_VDEVBITS 24 #define SPA_COMPRESSMASK ((1U << SPA_COMPRESSBITS) - 1) /* * 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; /* * Some checksums/hashes need a 256-bit initialization salt. This salt is kept * secret and is suitable for use in MAC algorithms as the key. */ typedef struct zio_cksum_salt { uint8_t zcs_bytes[32]; } zio_cksum_salt_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 | pad | vdev1 | GRID | ASIZE | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 1 |G| offset1 | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 2 | pad | vdev2 | GRID | ASIZE | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 3 |G| offset2 | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 4 | pad | 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 * E blkptr_t contains embedded data (see below) * lvl level of indirection * type DMU object type * phys birth txg when dva[0] was written; zero if same as logical birth txg * note that typically all the dva's would be written in this * txg, but they could be different if they were moved by * device removal. * 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 */ /* * The blkptr_t's of encrypted blocks also need to store the encryption * parameters so that the block can be decrypted. This layout 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 | salt | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 5 | IV1 | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 6 |BDX|lvl| type | cksum |E| comp| PSIZE | LSIZE | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 7 | padding | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 8 | padding | * +-------+-------+-------+-------+-------+-------+-------+-------+ * 9 | physical birth txg | * +-------+-------+-------+-------+-------+-------+-------+-------+ * a | logical birth txg | * +-------+-------+-------+-------+-------+-------+-------+-------+ * b | IV2 | fill count | * +-------+-------+-------+-------+-------+-------+-------+-------+ * c | checksum[0] | * +-------+-------+-------+-------+-------+-------+-------+-------+ * d | checksum[1] | * +-------+-------+-------+-------+-------+-------+-------+-------+ * e | MAC[0] | * +-------+-------+-------+-------+-------+-------+-------+-------+ * f | MAC[1] | * +-------+-------+-------+-------+-------+-------+-------+-------+ * * Legend: * * salt Salt for generating encryption keys * IV1 First 64 bits of encryption IV * X Block requires encryption handling (set to 1) * E blkptr_t contains embedded data (set to 0, see below) * fill count number of non-zero blocks under this bp (truncated to 32 bits) * IV2 Last 32 bits of encryption IV * checksum[2] 128-bit checksum of the data this bp describes * MAC[2] 128-bit message authentication code for this data * * The X bit being set indicates that this block is one of 3 types. If this is * a level 0 block with an encrypted object type, the block is encrypted * (see BP_IS_ENCRYPTED()). If this is a level 0 block with an unencrypted * object type, this block is authenticated with an HMAC (see * BP_IS_AUTHENTICATED()). Otherwise (if level > 0), this bp will use the MAC * words to store a checksum-of-MACs from the level below (see * BP_HAS_INDIRECT_MAC_CKSUM()). For convenience in the code, BP_IS_PROTECTED() * refers to both encrypted and authenticated blocks and BP_USES_CRYPT() * refers to any of these 3 kinds of blocks. * * The additional encryption parameters are the salt, IV, and MAC which are * explained in greater detail in the block comment at the top of zio_crypt.c. * The MAC occupies half of the checksum space since it serves a very similar * purpose: to prevent data corruption on disk. The only functional difference * is that the checksum is used to detect on-disk corruption whether or not the * encryption key is loaded and the MAC provides additional protection against * malicious disk tampering. We use the 3rd DVA to store the salt and first * 64 bits of the IV. As a result encrypted blocks can only have 2 copies * maximum instead of the normal 3. The last 32 bits of the IV are stored in * the upper bits of what is usually the fill count. Note that only blocks at * level 0 or -2 are ever encrypted, which allows us to guarantee that these * 32 bits are not trampled over by other code (see zio_crypt.c for details). * The salt and IV are not used for authenticated bps or bps with an indirect * MAC checksum, so these blocks can utilize all 3 DVAs and the full 64 bits * for the fill count. */ /* * "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) * 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". Encrypted blkptr_t's cannot be embedded because they use * the payload space for encryption parameters (see the comment above on * how encryption parameters are stored). */ #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); \ } 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); \ } 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); \ } while (0) typedef enum bp_embedded_type { BP_EMBEDDED_TYPE_DATA, BP_EMBEDDED_TYPE_RESERVED, /* Reserved for Delphix byteswap feature. */ BP_EMBEDDED_TYPE_REDACTED, NUM_BP_EMBEDDED_TYPES } 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 */ #define SPA_SYNC_MIN_VDEVS 3 /* min vdevs to update during sync */ /* * 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. */ /* * Note, for gang blocks, DVA_GET_ASIZE() is the total space allocated for * this gang DVA including its children BP's. The space allocated at this * DVA's vdev/offset is vdev_gang_header_asize(vdev). */ #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, SPA_VDEVBITS) #define DVA_SET_VDEV(dva, x) \ BF64_SET((dva)->dva_word[0], 32, SPA_VDEVBITS, 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); \ } 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); \ } while (0) #define BP_GET_COMPRESS(bp) \ BF64_GET((bp)->blk_prop, 32, SPA_COMPRESSBITS) #define BP_SET_COMPRESS(bp, x) \ BF64_SET((bp)->blk_prop, 32, SPA_COMPRESSBITS, 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); \ } 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) /* encrypted, authenticated, and MAC cksum bps use the same bit */ #define BP_USES_CRYPT(bp) BF64_GET((bp)->blk_prop, 61, 1) #define BP_SET_CRYPT(bp, x) BF64_SET((bp)->blk_prop, 61, 1, x) #define BP_IS_ENCRYPTED(bp) \ (BP_USES_CRYPT(bp) && \ BP_GET_LEVEL(bp) <= 0 && \ DMU_OT_IS_ENCRYPTED(BP_GET_TYPE(bp))) #define BP_IS_AUTHENTICATED(bp) \ (BP_USES_CRYPT(bp) && \ BP_GET_LEVEL(bp) <= 0 && \ !DMU_OT_IS_ENCRYPTED(BP_GET_TYPE(bp))) #define BP_HAS_INDIRECT_MAC_CKSUM(bp) \ (BP_USES_CRYPT(bp) && BP_GET_LEVEL(bp) > 0) #define BP_IS_PROTECTED(bp) \ (BP_IS_ENCRYPTED(bp) || BP_IS_AUTHENTICATED(bp)) #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_GET_FREE(bp) BF64_GET((bp)->blk_fill, 0, 1) #define BP_SET_FREE(bp, x) BF64_SET((bp)->blk_fill, 0, 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_ENCRYPTED(bp)) ? BF64_GET((bp)->blk_fill, 0, 32) : \ ((BP_IS_EMBEDDED(bp)) ? 1 : (bp)->blk_fill)) #define BP_SET_FILL(bp, fill) \ { \ if (BP_IS_ENCRYPTED(bp)) \ BF64_SET((bp)->blk_fill, 0, 32, fill); \ else \ (bp)->blk_fill = fill; \ } #define BP_GET_IV2(bp) \ (ASSERT(BP_IS_ENCRYPTED(bp)), \ BF64_GET((bp)->blk_fill, 32, 32)) #define BP_SET_IV2(bp, iv2) \ { \ ASSERT(BP_IS_ENCRYPTED(bp)); \ BF64_SET((bp)->blk_fill, 32, 32, iv2); \ } #define BP_IS_METADATA(bp) \ (BP_GET_LEVEL(bp) > 0 || DMU_OT_IS_METADATA(BP_GET_TYPE(bp))) #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]) * !BP_IS_ENCRYPTED(bp))) #define BP_GET_UCSIZE(bp) \ (BP_IS_METADATA(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]) * !BP_IS_ENCRYPTED(bp))) #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]) * !BP_IS_ENCRYPTED(bp)))) #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 DVA_IS_VALID(dva) (DVA_GET_ASIZE(dva) != 0) #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))) #define BP_SET_REDACTED(bp) \ { \ BP_SET_EMBEDDED(bp, B_TRUE); \ BPE_SET_ETYPE(bp, BP_EMBEDDED_TYPE_REDACTED); \ } #define BP_IS_REDACTED(bp) \ (BP_IS_EMBEDDED(bp) && BPE_GET_ETYPE(bp) == BP_EMBEDDED_TYPE_REDACTED) /* 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); \ } #ifdef _ZFS_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 400 /* * This macro allows code sharing between zfs, libzpool, and mdb. * 'func' is either kmem_scnprintf() 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 *const copyname[] = \ { "zero", "single", "double", "triple" }; \ int len = 0; \ int copies = 0; \ const char *crypt_type; \ if (bp != NULL) { \ if (BP_IS_ENCRYPTED(bp)) { \ crypt_type = "encrypted"; \ /* LINTED E_SUSPICIOUS_COMPARISON */ \ } else if (BP_IS_AUTHENTICATED(bp)) { \ crypt_type = "authenticated"; \ } else if (BP_HAS_INDIRECT_MAC_CKSUM(bp)) { \ crypt_type = "indirect-MAC"; \ } else { \ crypt_type = "unencrypted"; \ } \ } \ if (bp == NULL) { \ len += func(buf + len, size - len, ""); \ } else if (BP_IS_HOLE(bp)) { \ len += func(buf + len, size - len, \ "HOLE [L%llu %s] " \ "size=%llxL birth=%lluL", \ (u_longlong_t)BP_GET_LEVEL(bp), \ type, \ (u_longlong_t)BP_GET_LSIZE(bp), \ (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 if (BP_IS_REDACTED(bp)) { \ len += func(buf + len, size - len, \ "REDACTED [L%llu %s] size=%llxL birth=%lluL", \ (u_longlong_t)BP_GET_LEVEL(bp), \ type, \ (u_longlong_t)BP_GET_LSIZE(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); \ } \ ASSERT3S(copies, >, 0); \ if (BP_IS_ENCRYPTED(bp)) { \ len += func(buf + len, size - len, \ "salt=%llx iv=%llx:%llx%c", \ (u_longlong_t)bp->blk_dva[2].dva_word[0], \ (u_longlong_t)bp->blk_dva[2].dva_word[1], \ (u_longlong_t)BP_GET_IV2(bp), \ 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 %s%c" \ "size=%llxL/%llxP birth=%lluL/%lluP fill=%llu%c" \ "cksum=%016llx:%016llx:%016llx:%016llx", \ (u_longlong_t)BP_GET_LEVEL(bp), \ type, \ checksum, \ compress, \ crypt_type, \ 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); \ } #define BP_GET_BUFC_TYPE(bp) \ (BP_IS_METADATA(bp) ? ARC_BUFC_METADATA : ARC_BUFC_DATA) typedef enum spa_import_type { SPA_IMPORT_EXISTING, SPA_IMPORT_ASSEMBLE } spa_import_type_t; typedef enum spa_mode { SPA_MODE_UNINIT = 0, SPA_MODE_READ = 1, SPA_MODE_WRITE = 2, } spa_mode_t; /* * Send TRIM commands in-line during normal pool operation while deleting. * OFF: no * ON: yes * NB: IN_FREEBSD_BASE is defined within the FreeBSD sources. */ typedef enum { SPA_AUTOTRIM_OFF = 0, /* default */ SPA_AUTOTRIM_ON, #ifdef IN_FREEBSD_BASE SPA_AUTOTRIM_DEFAULT = SPA_AUTOTRIM_ON, #else SPA_AUTOTRIM_DEFAULT = SPA_AUTOTRIM_OFF, #endif } spa_autotrim_t; /* * Reason TRIM command was issued, used internally for accounting purposes. */ typedef enum trim_type { TRIM_TYPE_MANUAL = 0, TRIM_TYPE_AUTO = 1, TRIM_TYPE_SIMPLE = 2 } trim_type_t; /* state manipulation functions */ extern int spa_open(const char *pool, spa_t **, const void *tag); extern int spa_open_rewind(const char *pool, spa_t **, const 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 *nvroot, nvlist_t *props, nvlist_t *zplprops, struct dsl_crypto_params *dcp); extern int spa_import(char *pool, nvlist_t *config, nvlist_t *props, uint64_t flags); extern nvlist_t *spa_tryimport(nvlist_t *tryconfig); extern int spa_destroy(const char *pool); extern int spa_checkpoint(const char *pool); extern int spa_checkpoint_discard(const char *pool); extern int spa_export(const char *pool, nvlist_t **oldconfig, boolean_t force, boolean_t hardforce); extern int spa_reset(const 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 int spa_async_tasks(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); extern int bpobj_enqueue_alloc_cb(void *arg, const blkptr_t *bp, dmu_tx_t *tx); extern int bpobj_enqueue_free_cb(void *arg, const blkptr_t *bp, dmu_tx_t *tx); #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 #define SPA_ASYNC_INITIALIZE_RESTART 0x100 #define SPA_ASYNC_TRIM_RESTART 0x200 #define SPA_ASYNC_AUTOTRIM_RESTART 0x400 #define SPA_ASYNC_L2CACHE_REBUILD 0x800 #define SPA_ASYNC_L2CACHE_TRIM 0x1000 #define SPA_ASYNC_REBUILD_DONE 0x2000 #define SPA_ASYNC_DETACH_SPARE 0x4000 /* 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, int rebuild); extern int spa_vdev_detach(spa_t *spa, uint64_t guid, uint64_t pguid, int replace_done); extern int spa_vdev_alloc(spa_t *spa, uint64_t guid); extern int spa_vdev_noalloc(spa_t *spa, uint64_t guid); extern boolean_t spa_vdev_remove_active(spa_t *spa); extern int spa_vdev_initialize(spa_t *spa, nvlist_t *nv, uint64_t cmd_type, nvlist_t *vdev_errlist); extern int spa_vdev_trim(spa_t *spa, nvlist_t *nv, uint64_t cmd_type, uint64_t rate, boolean_t partial, boolean_t secure, nvlist_t *vdev_errlist); 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, const 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); extern int spa_scrub_pause_resume(spa_t *spa, pool_scrub_cmd_t flag); /* spa syncing */ extern void spa_sync(spa_t *spa, uint64_t txg); /* only for DMU use */ extern void spa_sync_allpools(void); extern uint_t zfs_sync_pass_deferred_free; /* 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_write_cachefile(spa_t *, boolean_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); extern int spa_config_parse(spa_t *spa, vdev_t **vdp, nvlist_t *nv, vdev_t *parent, uint_t id, int atype); /* * 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, const void *tag); extern void spa_close(spa_t *spa, const void *tag); extern void spa_async_close(spa_t *spa, const 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) /* Historical pool statistics */ typedef struct spa_history_kstat { kmutex_t lock; uint64_t count; uint64_t size; kstat_t *kstat; void *priv; list_t list; } spa_history_kstat_t; typedef struct spa_history_list { uint64_t size; procfs_list_t procfs_list; } spa_history_list_t; typedef struct spa_stats { spa_history_list_t read_history; spa_history_list_t txg_history; spa_history_kstat_t tx_assign_histogram; spa_history_list_t mmp_history; spa_history_kstat_t state; /* pool state */ spa_history_kstat_t guid; /* pool guid */ spa_history_kstat_t iostats; } spa_stats_t; typedef enum txg_state { TXG_STATE_BIRTH = 0, TXG_STATE_OPEN = 1, TXG_STATE_QUIESCED = 2, TXG_STATE_WAIT_FOR_SYNC = 3, TXG_STATE_SYNCED = 4, TXG_STATE_COMMITTED = 5, } txg_state_t; typedef struct txg_stat { vdev_stat_t vs1; vdev_stat_t vs2; uint64_t txg; uint64_t ndirty; } txg_stat_t; /* Assorted pool IO kstats */ typedef struct spa_iostats { kstat_named_t trim_extents_written; kstat_named_t trim_bytes_written; kstat_named_t trim_extents_skipped; kstat_named_t trim_bytes_skipped; kstat_named_t trim_extents_failed; kstat_named_t trim_bytes_failed; kstat_named_t autotrim_extents_written; kstat_named_t autotrim_bytes_written; kstat_named_t autotrim_extents_skipped; kstat_named_t autotrim_bytes_skipped; kstat_named_t autotrim_extents_failed; kstat_named_t autotrim_bytes_failed; kstat_named_t simple_trim_extents_written; kstat_named_t simple_trim_bytes_written; kstat_named_t simple_trim_extents_skipped; kstat_named_t simple_trim_bytes_skipped; kstat_named_t simple_trim_extents_failed; kstat_named_t simple_trim_bytes_failed; } spa_iostats_t; extern void spa_stats_init(spa_t *spa); extern void spa_stats_destroy(spa_t *spa); extern void spa_read_history_add(spa_t *spa, const zbookmark_phys_t *zb, uint32_t aflags); extern void spa_txg_history_add(spa_t *spa, uint64_t txg, hrtime_t birth_time); extern int spa_txg_history_set(spa_t *spa, uint64_t txg, txg_state_t completed_state, hrtime_t completed_time); extern txg_stat_t *spa_txg_history_init_io(spa_t *, uint64_t, struct dsl_pool *); extern void spa_txg_history_fini_io(spa_t *, txg_stat_t *); extern void spa_tx_assign_add_nsecs(spa_t *spa, uint64_t nsecs); extern int spa_mmp_history_set_skip(spa_t *spa, uint64_t mmp_kstat_id); extern int spa_mmp_history_set(spa_t *spa, uint64_t mmp_kstat_id, int io_error, hrtime_t duration); extern void spa_mmp_history_add(spa_t *spa, uint64_t txg, uint64_t timestamp, uint64_t mmp_delay, vdev_t *vd, int label, uint64_t mmp_kstat_id, int error); extern void spa_iostats_trim_add(spa_t *spa, trim_type_t type, uint64_t extents_written, uint64_t bytes_written, uint64_t extents_skipped, uint64_t bytes_skipped, uint64_t extents_failed, uint64_t bytes_failed); extern void spa_import_progress_add(spa_t *spa); extern void spa_import_progress_remove(uint64_t spa_guid); extern int spa_import_progress_set_mmp_check(uint64_t pool_guid, uint64_t mmp_sec_remaining); extern int spa_import_progress_set_max_txg(uint64_t pool_guid, uint64_t max_txg); extern int spa_import_progress_set_state(uint64_t pool_guid, spa_load_state_t spa_load_state); /* Pool configuration locks */ extern int spa_config_tryenter(spa_t *spa, int locks, const void *tag, krw_t rw); extern void spa_config_enter(spa_t *spa, int locks, const void *tag, krw_t rw); extern void spa_config_enter_mmp(spa_t *spa, int locks, const void *tag, krw_t rw); extern void spa_config_exit(spa_t *spa, int locks, const 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_detach_enter(spa_t *spa, uint64_t guid); 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, const 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_reset_logs(spa_t *spa); /* Log claim callback */ extern void spa_claim_notify(zio_t *zio); extern void spa_deadman(void *); /* 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 boolean_t spa_indirect_vdevs_loaded(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 uint32_t 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_final_dirty_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_worst_case_asize(spa_t *spa, uint64_t lsize); extern uint64_t spa_get_dspace(spa_t *spa); extern uint64_t spa_get_checkpoint_space(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 metaslab_class_t *spa_embedded_log_class(spa_t *spa); extern metaslab_class_t *spa_special_class(spa_t *spa); extern metaslab_class_t *spa_dedup_class(spa_t *spa); extern metaslab_class_t *spa_preferred_class(spa_t *spa, uint64_t size, dmu_object_type_t objtype, uint_t level, uint_t special_smallblk); 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 uint64_t spa_get_failmode(spa_t *spa); extern uint64_t spa_get_deadman_failmode(spa_t *spa); extern void spa_set_deadman_failmode(spa_t *spa, const char *failmode); 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 space_map_t *spa_syncing_log_sm(spa_t *spa); extern uint64_t spa_deadman_synctime(spa_t *spa); extern uint64_t spa_deadman_ziotime(spa_t *spa); extern uint64_t spa_dirty_data(spa_t *spa); extern spa_autotrim_t spa_get_autotrim(spa_t *spa); /* Miscellaneous support routines */ extern void spa_load_failed(spa_t *spa, const char *fmt, ...) __attribute__((format(printf, 2, 3))); extern void spa_load_note(spa_t *spa, const char *fmt, ...) __attribute__((format(printf, 2, 3))); 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 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_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_l2cache(spa_t *, uint64_t guid); 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 int spa_maxdnodesize(spa_t *spa); extern boolean_t spa_has_checkpoint(spa_t *spa); extern boolean_t spa_importing_readonly_checkpoint(spa_t *spa); extern boolean_t spa_suspend_async_destroy(spa_t *spa); extern uint64_t spa_min_claim_txg(spa_t *spa); extern boolean_t zfs_dva_valid(spa_t *spa, const dva_t *dva, const blkptr_t *bp); typedef void (*spa_remap_cb_t)(uint64_t vdev, uint64_t offset, uint64_t size, void *arg); extern boolean_t spa_remap_blkptr(spa_t *spa, blkptr_t *bp, spa_remap_cb_t callback, void *arg); extern uint64_t spa_get_last_removal_txg(spa_t *spa); extern boolean_t spa_trust_config(spa_t *spa); extern uint64_t spa_missing_tvds_allowed(spa_t *spa); extern void spa_set_missing_tvds(spa_t *spa, uint64_t missing); extern boolean_t spa_top_vdevs_spacemap_addressable(spa_t *spa); extern uint64_t spa_total_metaslabs(spa_t *spa); extern boolean_t spa_multihost(spa_t *spa); extern uint32_t spa_get_hostid(spa_t *spa); extern void spa_activate_allocation_classes(spa_t *, dmu_tx_t *); extern boolean_t spa_livelist_delete_check(spa_t *spa); extern spa_mode_t spa_mode(spa_t *spa); extern uint64_t zfs_strtonum(const char *str, char **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, dmu_tx_t *tx); extern void spa_history_log_internal(spa_t *spa, const char *operation, dmu_tx_t *tx, const char *fmt, ...) __printflike(4, 5); extern void spa_history_log_internal_ds(struct dsl_dataset *ds, const char *op, dmu_tx_t *tx, const char *fmt, ...) __printflike(4, 5); extern void spa_history_log_internal_dd(dsl_dir_t *dd, const char *operation, dmu_tx_t *tx, const char *fmt, ...) __printflike(4, 5); extern const char *spa_state_to_name(spa_t *spa); /* error handling */ struct zbookmark_phys; extern void spa_log_error(spa_t *spa, const zbookmark_phys_t *zb, const uint64_t *birth); -extern void spa_remove_error(spa_t *spa, zbookmark_phys_t *zb); +extern void spa_remove_error(spa_t *spa, zbookmark_phys_t *zb, + const uint64_t *birth); extern int zfs_ereport_post(const char *clazz, spa_t *spa, vdev_t *vd, const zbookmark_phys_t *zb, zio_t *zio, uint64_t state); extern boolean_t zfs_ereport_is_valid(const char *clazz, spa_t *spa, vdev_t *vd, zio_t *zio); extern void zfs_ereport_taskq_fini(void); extern void zfs_ereport_clear(spa_t *spa, vdev_t *vd); extern nvlist_t *zfs_event_create(spa_t *spa, vdev_t *vd, const char *type, const char *name, nvlist_t *aux); extern void zfs_post_remove(spa_t *spa, vdev_t *vd); extern void zfs_post_state_change(spa_t *spa, vdev_t *vd, uint64_t laststate); extern void zfs_post_autoreplace(spa_t *spa, vdev_t *vd); extern uint64_t spa_approx_errlog_size(spa_t *spa); extern int spa_get_errlog(spa_t *spa, void *uaddr, uint64_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); extern void spa_delete_dataset_errlog(spa_t *spa, uint64_t ds, dmu_tx_t *tx); extern void spa_swap_errlog(spa_t *spa, uint64_t new_head_ds, uint64_t old_head_ds, dmu_tx_t *tx); extern void sync_error_list(spa_t *spa, avl_tree_t *t, uint64_t *obj, dmu_tx_t *tx); extern void spa_upgrade_errlog(spa_t *spa, dmu_tx_t *tx); /* vdev cache */ extern void vdev_cache_stat_init(void); extern void vdev_cache_stat_fini(void); /* vdev mirror */ extern void vdev_mirror_stat_init(void); extern void vdev_mirror_stat_fini(void); /* Initialization and termination */ extern void spa_init(spa_mode_t mode); extern void spa_fini(void); extern void spa_boot_init(void); /* 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, nvlist_t *hist_nvl, const char *name); extern void zfs_ereport_zvol_post(const char *subclass, const char *name, const char *device_name, const char *raw_name); /* waiting for pool activities to complete */ extern int spa_wait(const char *pool, zpool_wait_activity_t activity, boolean_t *waited); extern int spa_wait_tag(const char *name, zpool_wait_activity_t activity, uint64_t tag, boolean_t *waited); extern void spa_notify_waiters(spa_t *spa); extern void spa_wake_waiters(spa_t *spa); extern void spa_import_os(spa_t *spa); extern void spa_export_os(spa_t *spa); extern void spa_activate_os(spa_t *spa); extern void spa_deactivate_os(spa_t *spa); /* module param call functions */ int param_set_deadman_ziotime(ZFS_MODULE_PARAM_ARGS); int param_set_deadman_synctime(ZFS_MODULE_PARAM_ARGS); int param_set_slop_shift(ZFS_MODULE_PARAM_ARGS); int param_set_deadman_failmode(ZFS_MODULE_PARAM_ARGS); #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); \ } \ } while (0) #else #define dprintf_bp(bp, fmt, ...) #endif extern spa_mode_t spa_mode_global; extern int zfs_deadman_enabled; extern uint64_t zfs_deadman_synctime_ms; extern uint64_t zfs_deadman_ziotime_ms; extern uint64_t zfs_deadman_checktime_ms; extern kmem_cache_t *zio_buf_cache[]; extern kmem_cache_t *zio_data_buf_cache[]; #ifdef __cplusplus } #endif #endif /* _SYS_SPA_H */ diff --git a/man/man8/zpool-status.8 b/man/man8/zpool-status.8 index ed572e29f51f..8f9580cf086e 100644 --- a/man/man8/zpool-status.8 +++ b/man/man8/zpool-status.8 @@ -1,162 +1,165 @@ .\" .\" 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 https://opensource.org/licenses/CDDL-1.0. .\" 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) 2007, Sun Microsystems, Inc. All Rights Reserved. .\" Copyright (c) 2012, 2018 by Delphix. All rights reserved. .\" Copyright (c) 2012 Cyril Plisko. All Rights Reserved. .\" Copyright (c) 2017 Datto Inc. .\" Copyright (c) 2018 George Melikov. All Rights Reserved. .\" Copyright 2017 Nexenta Systems, Inc. .\" Copyright (c) 2017 Open-E, Inc. All Rights Reserved. .\" .Dd March 16, 2022 .Dt ZPOOL-STATUS 8 .Os . .Sh NAME .Nm zpool-status .Nd show detailed health status for ZFS storage pools .Sh SYNOPSIS .Nm zpool .Cm status .Op Fl DigLpPstvx .Op Fl T Sy u Ns | Ns Sy d .Op Fl c Op Ar SCRIPT1 Ns Oo , Ns Ar SCRIPT2 Oc Ns … .Oo Ar pool Oc Ns … .Op Ar interval Op Ar count . .Sh DESCRIPTION Displays the detailed health status for the given pools. If no .Ar pool is specified, then the status of each pool in the system is displayed. For more information on pool and device health, see the .Sx Device Failure and Recovery section of .Xr zpoolconcepts 7 . .Pp If a scrub or resilver is in progress, this command reports the percentage done and the estimated time to completion. Both of these are only approximate, because the amount of data in the pool and the other workloads on the system can change. .Bl -tag -width Ds .It Fl c Op Ar SCRIPT1 Ns Oo , Ns Ar SCRIPT2 Oc Ns … Run a script (or scripts) on each vdev and include the output as a new column in the .Nm zpool Cm status output. See the .Fl c option of .Nm zpool Cm iostat for complete details. .It Fl i Display vdev initialization status. .It Fl g Display vdev GUIDs instead of the normal device names These GUIDs can be used in place of device names for the zpool detach/offline/remove/replace commands. .It Fl L Display real paths for vdevs resolving all symbolic links. This can be used to look up the current block device name regardless of the .Pa /dev/disk/ path used to open it. .It Fl p Display numbers in parsable (exact) values. .It Fl P Display full paths for vdevs instead of only the last component of the path. This can be used in conjunction with the .Fl L flag. .It Fl D Display a histogram of deduplication statistics, showing the allocated .Pq physically present on disk and referenced .Pq logically referenced in the pool block counts and sizes by reference count. .It Fl s Display the number of leaf vdev slow I/O operations. This is the number of I/O operations that didn't complete in .Sy zio_slow_io_ms milliseconds .Pq Sy 30000 No by default . This does not necessarily mean the I/O operations failed to complete, just took an unreasonably long amount of time. This may indicate a problem with the underlying storage. .It Fl t Display vdev TRIM status. .It Fl T Sy u Ns | Ns Sy d Display a time stamp. Specify .Sy u for a printed representation of the internal representation of time. See .Xr time 2 . Specify .Sy d for standard date format. See .Xr date 1 . .It Fl v Displays verbose data error information, printing out a complete list of all data errors since the last complete pool scrub. +If the head_errlog feature is enabled and files containing errors have been +removed then the respective filenames will not be reported in subsequent runs +of this command. .It Fl x Only display status for pools that are exhibiting errors or are otherwise unavailable. Warnings about pools not using the latest on-disk format will not be included. .El . .Sh EXAMPLES .\" These are, respectively, examples 16 from zpool.8 .\" Make sure to update them bidirectionally .Ss Example 1 : No Adding output columns Additional columns can be added to the .Nm zpool Cm status No and Nm zpool Cm iostat No output with Fl c . .Bd -literal -compact -offset Ds .No # Nm zpool Cm status Fl c Pa vendor , Ns Pa model , Ns Pa size NAME STATE READ WRITE CKSUM vendor model size tank ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 U1 ONLINE 0 0 0 SEAGATE ST8000NM0075 7.3T U10 ONLINE 0 0 0 SEAGATE ST8000NM0075 7.3T U11 ONLINE 0 0 0 SEAGATE ST8000NM0075 7.3T U12 ONLINE 0 0 0 SEAGATE ST8000NM0075 7.3T U13 ONLINE 0 0 0 SEAGATE ST8000NM0075 7.3T U14 ONLINE 0 0 0 SEAGATE ST8000NM0075 7.3T .No # Nm zpool Cm iostat Fl vc Pa size capacity operations bandwidth pool alloc free read write read write size ---------- ----- ----- ----- ----- ----- ----- ---- rpool 14.6G 54.9G 4 55 250K 2.69M sda1 14.6G 54.9G 4 55 250K 2.69M 70G ---------- ----- ----- ----- ----- ----- ----- ---- .Ed . .Sh SEE ALSO .Xr zpool-events 8 , .Xr zpool-history 8 , .Xr zpool-iostat 8 , .Xr zpool-list 8 , .Xr zpool-resilver 8 , .Xr zpool-scrub 8 , .Xr zpool-wait 8 diff --git a/module/zfs/dmu_recv.c b/module/zfs/dmu_recv.c index c2ce5ce000ac..c22a95f8647f 100644 --- a/module/zfs/dmu_recv.c +++ b/module/zfs/dmu_recv.c @@ -1,3799 +1,3799 @@ /* * 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 https://opensource.org/licenses/CDDL-1.0. * 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, 2020 by Delphix. All rights reserved. * Copyright (c) 2014, Joyent, Inc. All rights reserved. * Copyright 2014 HybridCluster. All rights reserved. * Copyright (c) 2018, loli10K . All rights reserved. * Copyright (c) 2019, Klara Inc. * Copyright (c) 2019, Allan Jude * Copyright (c) 2019 Datto Inc. * Copyright (c) 2022 Axcient. */ #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 #include #include #include #include #include #include #ifdef _KERNEL #include #endif #include static uint_t zfs_recv_queue_length = SPA_MAXBLOCKSIZE; static uint_t zfs_recv_queue_ff = 20; static uint_t zfs_recv_write_batch_size = 1024 * 1024; static int zfs_recv_best_effort_corrective = 0; static const void *const dmu_recv_tag = "dmu_recv_tag"; const char *const recv_clone_name = "%recv"; typedef enum { ORNS_NO, ORNS_YES, ORNS_MAYBE } or_need_sync_t; static int receive_read_payload_and_next_header(dmu_recv_cookie_t *ra, int len, void *buf); struct receive_record_arg { dmu_replay_record_t header; void *payload; /* Pointer to a buffer containing the payload */ /* * If the record is a WRITE or SPILL, pointer to the abd containing the * payload. */ abd_t *abd; int payload_size; uint64_t bytes_read; /* bytes read from stream when record created */ boolean_t eos_marker; /* Marks the end of the stream */ bqueue_node_t node; }; struct receive_writer_arg { objset_t *os; boolean_t byteswap; bqueue_t q; /* * These three members are used to signal to the main thread when * we're done. */ kmutex_t mutex; kcondvar_t cv; boolean_t done; int err; const char *tofs; boolean_t heal; boolean_t resumable; boolean_t raw; /* DMU_BACKUP_FEATURE_RAW set */ boolean_t spill; /* DRR_FLAG_SPILL_BLOCK set */ boolean_t full; /* this is a full send stream */ uint64_t last_object; uint64_t last_offset; uint64_t max_object; /* highest object ID referenced in stream */ uint64_t bytes_read; /* bytes read when current record created */ list_t write_batch; /* Encryption parameters for the last received DRR_OBJECT_RANGE */ boolean_t or_crypt_params_present; uint64_t or_firstobj; uint64_t or_numslots; uint8_t or_salt[ZIO_DATA_SALT_LEN]; uint8_t or_iv[ZIO_DATA_IV_LEN]; uint8_t or_mac[ZIO_DATA_MAC_LEN]; boolean_t or_byteorder; zio_t *heal_pio; /* Keep track of DRR_FREEOBJECTS right after DRR_OBJECT_RANGE */ or_need_sync_t or_need_sync; }; typedef struct dmu_recv_begin_arg { const char *drba_origin; dmu_recv_cookie_t *drba_cookie; cred_t *drba_cred; proc_t *drba_proc; dsl_crypto_params_t *drba_dcp; } dmu_recv_begin_arg_t; static void byteswap_record(dmu_replay_record_t *drr) { #define DO64(X) (drr->drr_u.X = BSWAP_64(drr->drr_u.X)) #define DO32(X) (drr->drr_u.X = BSWAP_32(drr->drr_u.X)) drr->drr_type = BSWAP_32(drr->drr_type); drr->drr_payloadlen = BSWAP_32(drr->drr_payloadlen); switch (drr->drr_type) { case DRR_BEGIN: DO64(drr_begin.drr_magic); DO64(drr_begin.drr_versioninfo); DO64(drr_begin.drr_creation_time); DO32(drr_begin.drr_type); DO32(drr_begin.drr_flags); DO64(drr_begin.drr_toguid); DO64(drr_begin.drr_fromguid); break; case DRR_OBJECT: DO64(drr_object.drr_object); DO32(drr_object.drr_type); DO32(drr_object.drr_bonustype); DO32(drr_object.drr_blksz); DO32(drr_object.drr_bonuslen); DO32(drr_object.drr_raw_bonuslen); DO64(drr_object.drr_toguid); DO64(drr_object.drr_maxblkid); break; case DRR_FREEOBJECTS: DO64(drr_freeobjects.drr_firstobj); DO64(drr_freeobjects.drr_numobjs); DO64(drr_freeobjects.drr_toguid); break; case DRR_WRITE: DO64(drr_write.drr_object); DO32(drr_write.drr_type); DO64(drr_write.drr_offset); DO64(drr_write.drr_logical_size); DO64(drr_write.drr_toguid); ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_write.drr_key.ddk_cksum); DO64(drr_write.drr_key.ddk_prop); DO64(drr_write.drr_compressed_size); break; case DRR_WRITE_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); DO64(drr_spill.drr_compressed_size); DO32(drr_spill.drr_type); break; case DRR_OBJECT_RANGE: DO64(drr_object_range.drr_firstobj); DO64(drr_object_range.drr_numslots); DO64(drr_object_range.drr_toguid); break; case DRR_REDACT: DO64(drr_redact.drr_object); DO64(drr_redact.drr_offset); DO64(drr_redact.drr_length); DO64(drr_redact.drr_toguid); break; case DRR_END: DO64(drr_end.drr_toguid); ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_end.drr_checksum); break; default: break; } if (drr->drr_type != DRR_BEGIN) { ZIO_CHECKSUM_BSWAP(&drr->drr_u.drr_checksum.drr_checksum); } #undef DO64 #undef DO32 } static boolean_t redact_snaps_contains(uint64_t *snaps, uint64_t num_snaps, uint64_t guid) { for (int i = 0; i < num_snaps; i++) { if (snaps[i] == guid) return (B_TRUE); } return (B_FALSE); } /* * Check that the new stream we're trying to receive is redacted with respect to * a subset of the snapshots that the origin was redacted with respect to. For * the reasons behind this, see the man page on redacted zfs sends and receives. */ static boolean_t compatible_redact_snaps(uint64_t *origin_snaps, uint64_t origin_num_snaps, uint64_t *redact_snaps, uint64_t num_redact_snaps) { /* * Short circuit the comparison; if we are redacted with respect to * more snapshots than the origin, we can't be redacted with respect * to a subset. */ if (num_redact_snaps > origin_num_snaps) { return (B_FALSE); } for (int i = 0; i < num_redact_snaps; i++) { if (!redact_snaps_contains(origin_snaps, origin_num_snaps, redact_snaps[i])) { return (B_FALSE); } } return (B_TRUE); } static boolean_t redact_check(dmu_recv_begin_arg_t *drba, dsl_dataset_t *origin) { uint64_t *origin_snaps; uint64_t origin_num_snaps; dmu_recv_cookie_t *drc = drba->drba_cookie; struct drr_begin *drrb = drc->drc_drrb; int featureflags = DMU_GET_FEATUREFLAGS(drrb->drr_versioninfo); int err = 0; boolean_t ret = B_TRUE; uint64_t *redact_snaps; uint_t numredactsnaps; /* * If this is a full send stream, we're safe no matter what. */ if (drrb->drr_fromguid == 0) return (ret); VERIFY(dsl_dataset_get_uint64_array_feature(origin, SPA_FEATURE_REDACTED_DATASETS, &origin_num_snaps, &origin_snaps)); if (nvlist_lookup_uint64_array(drc->drc_begin_nvl, BEGINNV_REDACT_FROM_SNAPS, &redact_snaps, &numredactsnaps) == 0) { /* * If the send stream was sent from the redaction bookmark or * the redacted version of the dataset, then we're safe. Verify * that this is from the a compatible redaction bookmark or * redacted dataset. */ if (!compatible_redact_snaps(origin_snaps, origin_num_snaps, redact_snaps, numredactsnaps)) { err = EINVAL; } } else if (featureflags & DMU_BACKUP_FEATURE_REDACTED) { /* * If the stream is redacted, it must be redacted with respect * to a subset of what the origin is redacted with respect to. * See case number 2 in the zfs man page section on redacted zfs * send. */ err = nvlist_lookup_uint64_array(drc->drc_begin_nvl, BEGINNV_REDACT_SNAPS, &redact_snaps, &numredactsnaps); if (err != 0 || !compatible_redact_snaps(origin_snaps, origin_num_snaps, redact_snaps, numredactsnaps)) { err = EINVAL; } } else if (!redact_snaps_contains(origin_snaps, origin_num_snaps, drrb->drr_toguid)) { /* * If the stream isn't redacted but the origin is, this must be * one of the snapshots the origin is redacted with respect to. * See case number 1 in the zfs man page section on redacted zfs * send. */ err = EINVAL; } if (err != 0) ret = B_FALSE; return (ret); } /* * If we previously received a stream with --large-block, we don't support * receiving an incremental on top of it without --large-block. This avoids * forcing a read-modify-write or trying to re-aggregate a string of WRITE * records. */ static int recv_check_large_blocks(dsl_dataset_t *ds, uint64_t featureflags) { if (dsl_dataset_feature_is_active(ds, SPA_FEATURE_LARGE_BLOCKS) && !(featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS)) return (SET_ERROR(ZFS_ERR_STREAM_LARGE_BLOCK_MISMATCH)); return (0); } static int recv_begin_check_existing_impl(dmu_recv_begin_arg_t *drba, dsl_dataset_t *ds, uint64_t fromguid, uint64_t featureflags) { uint64_t obj; uint64_t children; int error; dsl_dataset_t *snap; dsl_pool_t *dp = ds->ds_dir->dd_pool; boolean_t encrypted = ds->ds_dir->dd_crypto_obj != 0; boolean_t raw = (featureflags & DMU_BACKUP_FEATURE_RAW) != 0; boolean_t embed = (featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) != 0; /* 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, &obj); if (error != ENOENT) return (error == 0 ? SET_ERROR(EBUSY) : error); /* Resume state must not be set. */ if (dsl_dataset_has_resume_receive_state(ds)) return (SET_ERROR(EBUSY)); /* New snapshot name must not exist if we're not healing it. */ error = zap_lookup(dp->dp_meta_objset, dsl_dataset_phys(ds)->ds_snapnames_zapobj, drba->drba_cookie->drc_tosnap, 8, 1, &obj); if (drba->drba_cookie->drc_heal) { if (error != 0) return (error); } else if (error != ENOENT) { return (error == 0 ? SET_ERROR(EEXIST) : error); } /* Must not have children if receiving a ZVOL. */ error = zap_count(dp->dp_meta_objset, dsl_dir_phys(ds->ds_dir)->dd_child_dir_zapobj, &children); if (error != 0) return (error); if (drba->drba_cookie->drc_drrb->drr_type != DMU_OST_ZFS && children > 0) return (SET_ERROR(ZFS_ERR_WRONG_PARENT)); /* * 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, drba->drba_proc); if (error != 0) return (error); if (drba->drba_cookie->drc_heal) { /* Encryption is incompatible with embedded data. */ if (encrypted && embed) return (SET_ERROR(EINVAL)); /* Healing is not supported when in 'force' mode. */ if (drba->drba_cookie->drc_force) return (SET_ERROR(EINVAL)); /* Must have keys loaded if doing encrypted non-raw recv. */ if (encrypted && !raw) { if (spa_keystore_lookup_key(dp->dp_spa, ds->ds_object, NULL, NULL) != 0) return (SET_ERROR(EACCES)); } error = dsl_dataset_hold_obj(dp, obj, FTAG, &snap); if (error != 0) return (error); /* * When not doing best effort corrective recv healing can only * be done if the send stream is for the same snapshot as the * one we are trying to heal. */ if (zfs_recv_best_effort_corrective == 0 && drba->drba_cookie->drc_drrb->drr_toguid != dsl_dataset_phys(snap)->ds_guid) { dsl_dataset_rele(snap, FTAG); return (SET_ERROR(ENOTSUP)); } dsl_dataset_rele(snap, FTAG); } else if (fromguid != 0) { /* Sanity check the incremental recv */ uint64_t obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; /* Can't perform a raw receive on top of a non-raw receive */ if (!encrypted && raw) return (SET_ERROR(EINVAL)); /* Encryption is incompatible with embedded data */ if (encrypted && embed) return (SET_ERROR(EINVAL)); /* 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_cookie->drc_fromsnapobj = obj; } else { /* * If we are not forcing, there must be no * changes since fromsnap. Raw sends have an * additional constraint that requires that * no "noop" snapshots exist between fromsnap * and tosnap for the IVset checking code to * work properly. */ if (dsl_dataset_modified_since_snap(ds, snap) || (raw && dsl_dataset_phys(ds)->ds_prev_snap_obj != snap->ds_object)) { dsl_dataset_rele(snap, FTAG); return (SET_ERROR(ETXTBSY)); } drba->drba_cookie->drc_fromsnapobj = ds->ds_prev->ds_object; } if (dsl_dataset_feature_is_active(snap, SPA_FEATURE_REDACTED_DATASETS) && !redact_check(drba, snap)) { dsl_dataset_rele(snap, FTAG); return (SET_ERROR(EINVAL)); } error = recv_check_large_blocks(snap, featureflags); if (error != 0) { dsl_dataset_rele(snap, FTAG); return (error); } dsl_dataset_rele(snap, FTAG); } else { /* If full and not healing then must be forced. */ if (!drba->drba_cookie->drc_force) return (SET_ERROR(EEXIST)); /* * We don't support using zfs recv -F to blow away * encrypted filesystems. This would require the * dsl dir to point to the old encryption key and * the new one at the same time during the receive. */ if ((!encrypted && raw) || encrypted) return (SET_ERROR(EINVAL)); /* * Perform the same encryption checks we would if * we were creating a new dataset from scratch. */ if (!raw) { boolean_t will_encrypt; error = dmu_objset_create_crypt_check( ds->ds_dir->dd_parent, drba->drba_dcp, &will_encrypt); if (error != 0) return (error); if (will_encrypt && embed) return (SET_ERROR(EINVAL)); } } return (0); } /* * Check that any feature flags used in the data stream we're receiving are * supported by the pool we are receiving into. * * Note that some of the features we explicitly check here have additional * (implicit) features they depend on, but those dependencies are enforced * through the zfeature_register() calls declaring the features that we * explicitly check. */ static int recv_begin_check_feature_flags_impl(uint64_t featureflags, spa_t *spa) { /* * Check if there are any unsupported feature flags. */ if (!DMU_STREAM_SUPPORTED(featureflags)) { return (SET_ERROR(ZFS_ERR_UNKNOWN_SEND_STREAM_FEATURE)); } /* Verify pool version supports SA if SA_SPILL feature set */ if ((featureflags & DMU_BACKUP_FEATURE_SA_SPILL) && spa_version(spa) < SPA_VERSION_SA) return (SET_ERROR(ENOTSUP)); /* * LZ4 compressed, ZSTD compressed, embedded, mooched, large blocks, * and large_dnodes in the stream can only be used if those pool * features are enabled because we don't attempt to decompress / * un-embed / un-mooch / split up the blocks / dnodes during the * receive process. */ if ((featureflags & DMU_BACKUP_FEATURE_LZ4) && !spa_feature_is_enabled(spa, SPA_FEATURE_LZ4_COMPRESS)) return (SET_ERROR(ENOTSUP)); if ((featureflags & DMU_BACKUP_FEATURE_ZSTD) && !spa_feature_is_enabled(spa, SPA_FEATURE_ZSTD_COMPRESS)) return (SET_ERROR(ENOTSUP)); if ((featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) && !spa_feature_is_enabled(spa, SPA_FEATURE_EMBEDDED_DATA)) return (SET_ERROR(ENOTSUP)); if ((featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) && !spa_feature_is_enabled(spa, SPA_FEATURE_LARGE_BLOCKS)) return (SET_ERROR(ENOTSUP)); if ((featureflags & DMU_BACKUP_FEATURE_LARGE_DNODE) && !spa_feature_is_enabled(spa, SPA_FEATURE_LARGE_DNODE)) return (SET_ERROR(ENOTSUP)); /* * Receiving redacted streams requires that redacted datasets are * enabled. */ if ((featureflags & DMU_BACKUP_FEATURE_REDACTED) && !spa_feature_is_enabled(spa, SPA_FEATURE_REDACTED_DATASETS)) return (SET_ERROR(ENOTSUP)); 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; ds_hold_flags_t dsflags = DS_HOLD_FLAG_NONE; int error; uint64_t featureflags = drba->drba_cookie->drc_featureflags; dsl_dataset_t *ds; const char *tofs = drba->drba_cookie->drc_tofs; /* already checked */ ASSERT3U(drrb->drr_magic, ==, DMU_BACKUP_MAGIC); ASSERT(!(featureflags & DMU_BACKUP_FEATURE_RESUMING)); if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == DMU_COMPOUNDSTREAM || drrb->drr_type >= DMU_OST_NUMTYPES || ((flags & DRR_FLAG_CLONE) && drba->drba_origin == NULL)) return (SET_ERROR(EINVAL)); error = recv_begin_check_feature_flags_impl(featureflags, dp->dp_spa); if (error != 0) return (error); /* Resumable receives require extensible datasets */ if (drba->drba_cookie->drc_resumable && !spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_EXTENSIBLE_DATASET)) return (SET_ERROR(ENOTSUP)); if (featureflags & DMU_BACKUP_FEATURE_RAW) { /* raw receives require the encryption feature */ if (!spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_ENCRYPTION)) return (SET_ERROR(ENOTSUP)); /* embedded data is incompatible with encryption and raw recv */ if (featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) return (SET_ERROR(EINVAL)); /* raw receives require spill block allocation flag */ if (!(flags & DRR_FLAG_SPILL_BLOCK)) return (SET_ERROR(ZFS_ERR_SPILL_BLOCK_FLAG_MISSING)); } else { /* * We support unencrypted datasets below encrypted ones now, * so add the DS_HOLD_FLAG_DECRYPT flag only if we are dealing * with a dataset we may encrypt. */ if (drba->drba_dcp == NULL || drba->drba_dcp->cp_crypt != ZIO_CRYPT_OFF) { dsflags |= DS_HOLD_FLAG_DECRYPT; } } error = dsl_dataset_hold_flags(dp, tofs, dsflags, FTAG, &ds); if (error == 0) { /* target fs already exists; recv into temp clone */ /* Can't recv a clone into an existing fs */ if (flags & DRR_FLAG_CLONE || drba->drba_origin) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } error = recv_begin_check_existing_impl(drba, ds, fromguid, featureflags); dsl_dataset_rele_flags(ds, dsflags, FTAG); } else if (error == ENOENT) { /* target fs does not exist; must be a full backup or clone */ char buf[ZFS_MAX_DATASET_NAME_LEN]; objset_t *os; /* healing recv must be done "into" an existing snapshot */ if (drba->drba_cookie->drc_heal == B_TRUE) return (SET_ERROR(ENOTSUP)); /* * If it's a non-clone incremental, we are missing the * target fs, so fail the recv. */ if (fromguid != 0 && !((flags & DRR_FLAG_CLONE) || drba->drba_origin)) return (SET_ERROR(ENOENT)); /* * If we're receiving a full send as a clone, and it doesn't * contain all the necessary free records and freeobject * records, reject it. */ if (fromguid == 0 && drba->drba_origin != NULL && !(flags & DRR_FLAG_FREERECORDS)) return (SET_ERROR(EINVAL)); /* Open the parent of tofs */ ASSERT3U(strlen(tofs), <, sizeof (buf)); (void) strlcpy(buf, tofs, strrchr(tofs, '/') - tofs + 1); error = dsl_dataset_hold(dp, buf, FTAG, &ds); if (error != 0) return (error); if ((featureflags & DMU_BACKUP_FEATURE_RAW) == 0 && drba->drba_origin == NULL) { boolean_t will_encrypt; /* * Check that we aren't breaking any encryption rules * and that we have all the parameters we need to * create an encrypted dataset if necessary. If we are * making an encrypted dataset the stream can't have * embedded data. */ error = dmu_objset_create_crypt_check(ds->ds_dir, drba->drba_dcp, &will_encrypt); if (error != 0) { dsl_dataset_rele(ds, FTAG); return (error); } if (will_encrypt && (featureflags & DMU_BACKUP_FEATURE_EMBED_DATA)) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(EINVAL)); } } /* * 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, drba->drba_proc); 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, drba->drba_proc); if (error != 0) { dsl_dataset_rele(ds, FTAG); return (error); } /* can't recv below anything but filesystems (eg. no ZVOLs) */ error = dmu_objset_from_ds(ds, &os); if (error != 0) { dsl_dataset_rele(ds, FTAG); return (error); } if (dmu_objset_type(os) != DMU_OST_ZFS) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(ZFS_ERR_WRONG_PARENT)); } if (drba->drba_origin != NULL) { dsl_dataset_t *origin; error = dsl_dataset_hold_flags(dp, drba->drba_origin, dsflags, FTAG, &origin); if (error != 0) { dsl_dataset_rele(ds, FTAG); return (error); } if (!origin->ds_is_snapshot) { dsl_dataset_rele_flags(origin, dsflags, FTAG); dsl_dataset_rele(ds, FTAG); return (SET_ERROR(EINVAL)); } if (dsl_dataset_phys(origin)->ds_guid != fromguid && fromguid != 0) { dsl_dataset_rele_flags(origin, dsflags, FTAG); dsl_dataset_rele(ds, FTAG); return (SET_ERROR(ENODEV)); } if (origin->ds_dir->dd_crypto_obj != 0 && (featureflags & DMU_BACKUP_FEATURE_EMBED_DATA)) { dsl_dataset_rele_flags(origin, dsflags, FTAG); dsl_dataset_rele(ds, FTAG); return (SET_ERROR(EINVAL)); } /* * If the origin is redacted we need to verify that this * send stream can safely be received on top of the * origin. */ if (dsl_dataset_feature_is_active(origin, SPA_FEATURE_REDACTED_DATASETS)) { if (!redact_check(drba, origin)) { dsl_dataset_rele_flags(origin, dsflags, FTAG); dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } } error = recv_check_large_blocks(ds, featureflags); if (error != 0) { dsl_dataset_rele_flags(origin, dsflags, FTAG); dsl_dataset_rele_flags(ds, dsflags, FTAG); return (error); } dsl_dataset_rele_flags(origin, dsflags, FTAG); } dsl_dataset_rele(ds, FTAG); error = 0; } return (error); } static void dmu_recv_begin_sync(void *arg, dmu_tx_t *tx) { dmu_recv_begin_arg_t *drba = arg; dsl_pool_t *dp = dmu_tx_pool(tx); objset_t *mos = dp->dp_meta_objset; dmu_recv_cookie_t *drc = drba->drba_cookie; struct drr_begin *drrb = drc->drc_drrb; const char *tofs = drc->drc_tofs; uint64_t featureflags = drc->drc_featureflags; dsl_dataset_t *ds, *newds; objset_t *os; uint64_t dsobj; ds_hold_flags_t dsflags = DS_HOLD_FLAG_NONE; int error; uint64_t crflags = 0; dsl_crypto_params_t dummy_dcp = { 0 }; dsl_crypto_params_t *dcp = drba->drba_dcp; if (drrb->drr_flags & DRR_FLAG_CI_DATA) crflags |= DS_FLAG_CI_DATASET; if ((featureflags & DMU_BACKUP_FEATURE_RAW) == 0) dsflags |= DS_HOLD_FLAG_DECRYPT; /* * Raw, non-incremental recvs always use a dummy dcp with * the raw cmd set. Raw incremental recvs do not use a dcp * since the encryption parameters are already set in stone. */ if (dcp == NULL && drrb->drr_fromguid == 0 && drba->drba_origin == NULL) { ASSERT3P(dcp, ==, NULL); dcp = &dummy_dcp; if (featureflags & DMU_BACKUP_FEATURE_RAW) dcp->cp_cmd = DCP_CMD_RAW_RECV; } error = dsl_dataset_hold_flags(dp, tofs, dsflags, FTAG, &ds); if (error == 0) { /* Create temporary clone unless we're doing corrective recv */ dsl_dataset_t *snap = NULL; if (drba->drba_cookie->drc_fromsnapobj != 0) { VERIFY0(dsl_dataset_hold_obj(dp, drba->drba_cookie->drc_fromsnapobj, FTAG, &snap)); ASSERT3P(dcp, ==, NULL); } if (drc->drc_heal) { /* When healing we want to use the provided snapshot */ VERIFY0(dsl_dataset_snap_lookup(ds, drc->drc_tosnap, &dsobj)); } else { dsobj = dsl_dataset_create_sync(ds->ds_dir, recv_clone_name, snap, crflags, drba->drba_cred, dcp, tx); } if (drba->drba_cookie->drc_fromsnapobj != 0) dsl_dataset_rele(snap, FTAG); dsl_dataset_rele_flags(ds, dsflags, 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)); ASSERT3P(dcp, ==, NULL); } /* Create new dataset. */ dsobj = dsl_dataset_create_sync(dd, strrchr(tofs, '/') + 1, origin, crflags, drba->drba_cred, dcp, tx); if (origin != NULL) dsl_dataset_rele(origin, FTAG); dsl_dir_rele(dd, FTAG); drc->drc_newfs = B_TRUE; } VERIFY0(dsl_dataset_own_obj_force(dp, dsobj, dsflags, dmu_recv_tag, &newds)); if (dsl_dataset_feature_is_active(newds, SPA_FEATURE_REDACTED_DATASETS)) { /* * If the origin dataset is redacted, the child will be redacted * when we create it. We clear the new dataset's * redaction info; if it should be redacted, we'll fill * in its information later. */ dsl_dataset_deactivate_feature(newds, SPA_FEATURE_REDACTED_DATASETS, tx); } VERIFY0(dmu_objset_from_ds(newds, &os)); if (drc->drc_resumable) { dsl_dataset_zapify(newds, tx); if (drrb->drr_fromguid != 0) { VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_FROMGUID, 8, 1, &drrb->drr_fromguid, tx)); } VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_TOGUID, 8, 1, &drrb->drr_toguid, tx)); VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_TONAME, 1, strlen(drrb->drr_toname) + 1, drrb->drr_toname, tx)); uint64_t one = 1; uint64_t zero = 0; VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_OBJECT, 8, 1, &one, tx)); VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_OFFSET, 8, 1, &zero, tx)); VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_BYTES, 8, 1, &zero, tx)); if (featureflags & DMU_BACKUP_FEATURE_LARGE_BLOCKS) { VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_LARGEBLOCK, 8, 1, &one, tx)); } if (featureflags & DMU_BACKUP_FEATURE_EMBED_DATA) { VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_EMBEDOK, 8, 1, &one, tx)); } if (featureflags & DMU_BACKUP_FEATURE_COMPRESSED) { VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_COMPRESSOK, 8, 1, &one, tx)); } if (featureflags & DMU_BACKUP_FEATURE_RAW) { VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_RAWOK, 8, 1, &one, tx)); } uint64_t *redact_snaps; uint_t numredactsnaps; if (nvlist_lookup_uint64_array(drc->drc_begin_nvl, BEGINNV_REDACT_FROM_SNAPS, &redact_snaps, &numredactsnaps) == 0) { VERIFY0(zap_add(mos, dsobj, DS_FIELD_RESUME_REDACT_BOOKMARK_SNAPS, sizeof (*redact_snaps), numredactsnaps, redact_snaps, tx)); } } /* * Usually the os->os_encrypted value is tied to the presence of a * DSL Crypto Key object in the dd. However, that will not be received * until dmu_recv_stream(), so we set the value manually for now. */ if (featureflags & DMU_BACKUP_FEATURE_RAW) { os->os_encrypted = B_TRUE; drba->drba_cookie->drc_raw = B_TRUE; } if (featureflags & DMU_BACKUP_FEATURE_REDACTED) { uint64_t *redact_snaps; uint_t numredactsnaps; VERIFY0(nvlist_lookup_uint64_array(drc->drc_begin_nvl, BEGINNV_REDACT_SNAPS, &redact_snaps, &numredactsnaps)); dsl_dataset_activate_redaction(newds, redact_snaps, numredactsnaps, tx); } dmu_buf_will_dirty(newds->ds_dbuf, tx); dsl_dataset_phys(newds)->ds_flags |= DS_FLAG_INCONSISTENT; /* * If we actually created a non-clone, we need to create the objset * in our new dataset. If this is a raw send we postpone this until * dmu_recv_stream() so that we can allocate the metadnode with the * properties from the DRR_BEGIN payload. */ rrw_enter(&newds->ds_bp_rwlock, RW_READER, FTAG); if (BP_IS_HOLE(dsl_dataset_get_blkptr(newds)) && (featureflags & DMU_BACKUP_FEATURE_RAW) == 0 && !drc->drc_heal) { (void) dmu_objset_create_impl(dp->dp_spa, newds, dsl_dataset_get_blkptr(newds), drrb->drr_type, tx); } rrw_exit(&newds->ds_bp_rwlock, FTAG); drba->drba_cookie->drc_ds = newds; drba->drba_cookie->drc_os = os; spa_history_log_internal_ds(newds, "receive", tx, " "); } static int dmu_recv_resume_begin_check(void *arg, dmu_tx_t *tx) { dmu_recv_begin_arg_t *drba = arg; dmu_recv_cookie_t *drc = drba->drba_cookie; dsl_pool_t *dp = dmu_tx_pool(tx); struct drr_begin *drrb = drc->drc_drrb; int error; ds_hold_flags_t dsflags = DS_HOLD_FLAG_NONE; dsl_dataset_t *ds; const char *tofs = drc->drc_tofs; /* already checked */ ASSERT3U(drrb->drr_magic, ==, DMU_BACKUP_MAGIC); ASSERT(drc->drc_featureflags & DMU_BACKUP_FEATURE_RESUMING); if (DMU_GET_STREAM_HDRTYPE(drrb->drr_versioninfo) == DMU_COMPOUNDSTREAM || drrb->drr_type >= DMU_OST_NUMTYPES) return (SET_ERROR(EINVAL)); /* * This is mostly a sanity check since we should have already done these * checks during a previous attempt to receive the data. */ error = recv_begin_check_feature_flags_impl(drc->drc_featureflags, dp->dp_spa); if (error != 0) return (error); /* 6 extra bytes for /%recv */ char recvname[ZFS_MAX_DATASET_NAME_LEN + 6]; (void) snprintf(recvname, sizeof (recvname), "%s/%s", tofs, recv_clone_name); if (drc->drc_featureflags & DMU_BACKUP_FEATURE_RAW) { /* raw receives require spill block allocation flag */ if (!(drrb->drr_flags & DRR_FLAG_SPILL_BLOCK)) return (SET_ERROR(ZFS_ERR_SPILL_BLOCK_FLAG_MISSING)); } else { dsflags |= DS_HOLD_FLAG_DECRYPT; } boolean_t recvexist = B_TRUE; if (dsl_dataset_hold_flags(dp, recvname, dsflags, FTAG, &ds) != 0) { /* %recv does not exist; continue in tofs */ recvexist = B_FALSE; error = dsl_dataset_hold_flags(dp, tofs, dsflags, FTAG, &ds); if (error != 0) return (error); } /* * Resume of full/newfs recv on existing dataset should be done with * force flag */ if (recvexist && drrb->drr_fromguid == 0 && !drc->drc_force) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(ZFS_ERR_RESUME_EXISTS)); } /* check that ds is marked inconsistent */ if (!DS_IS_INCONSISTENT(ds)) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } /* check that there is resuming data, and that the toguid matches */ if (!dsl_dataset_is_zapified(ds)) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } uint64_t val; error = zap_lookup(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_TOGUID, sizeof (val), 1, &val); if (error != 0 || drrb->drr_toguid != val) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } /* * Check if the receive is still running. If so, it will be owned. * Note that nothing else can own the dataset (e.g. after the receive * fails) because it will be marked inconsistent. */ if (dsl_dataset_has_owner(ds)) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EBUSY)); } /* There should not be any snapshots of this fs yet. */ if (ds->ds_prev != NULL && ds->ds_prev->ds_dir == ds->ds_dir) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } /* * Note: resume point will be checked when we process the first WRITE * record. */ /* check that the origin matches */ val = 0; (void) zap_lookup(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_FROMGUID, sizeof (val), 1, &val); if (drrb->drr_fromguid != val) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } if (ds->ds_prev != NULL && drrb->drr_fromguid != 0) drc->drc_fromsnapobj = ds->ds_prev->ds_object; /* * If we're resuming, and the send is redacted, then the original send * must have been redacted, and must have been redacted with respect to * the same snapshots. */ if (drc->drc_featureflags & DMU_BACKUP_FEATURE_REDACTED) { uint64_t num_ds_redact_snaps; uint64_t *ds_redact_snaps; uint_t num_stream_redact_snaps; uint64_t *stream_redact_snaps; if (nvlist_lookup_uint64_array(drc->drc_begin_nvl, BEGINNV_REDACT_SNAPS, &stream_redact_snaps, &num_stream_redact_snaps) != 0) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } if (!dsl_dataset_get_uint64_array_feature(ds, SPA_FEATURE_REDACTED_DATASETS, &num_ds_redact_snaps, &ds_redact_snaps)) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } for (int i = 0; i < num_ds_redact_snaps; i++) { if (!redact_snaps_contains(ds_redact_snaps, num_ds_redact_snaps, stream_redact_snaps[i])) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (SET_ERROR(EINVAL)); } } } error = recv_check_large_blocks(ds, drc->drc_featureflags); if (error != 0) { dsl_dataset_rele_flags(ds, dsflags, FTAG); return (error); } dsl_dataset_rele_flags(ds, dsflags, FTAG); return (0); } static void dmu_recv_resume_begin_sync(void *arg, dmu_tx_t *tx) { dmu_recv_begin_arg_t *drba = arg; dsl_pool_t *dp = dmu_tx_pool(tx); const char *tofs = drba->drba_cookie->drc_tofs; uint64_t featureflags = drba->drba_cookie->drc_featureflags; dsl_dataset_t *ds; ds_hold_flags_t dsflags = DS_HOLD_FLAG_NONE; /* 6 extra bytes for /%recv */ char recvname[ZFS_MAX_DATASET_NAME_LEN + 6]; (void) snprintf(recvname, sizeof (recvname), "%s/%s", tofs, recv_clone_name); if (featureflags & DMU_BACKUP_FEATURE_RAW) { drba->drba_cookie->drc_raw = B_TRUE; } else { dsflags |= DS_HOLD_FLAG_DECRYPT; } if (dsl_dataset_own_force(dp, recvname, dsflags, dmu_recv_tag, &ds) != 0) { /* %recv does not exist; continue in tofs */ VERIFY0(dsl_dataset_own_force(dp, tofs, dsflags, dmu_recv_tag, &ds)); drba->drba_cookie->drc_newfs = B_TRUE; } ASSERT(DS_IS_INCONSISTENT(ds)); rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG); ASSERT(!BP_IS_HOLE(dsl_dataset_get_blkptr(ds)) || drba->drba_cookie->drc_raw); rrw_exit(&ds->ds_bp_rwlock, FTAG); drba->drba_cookie->drc_ds = ds; VERIFY0(dmu_objset_from_ds(ds, &drba->drba_cookie->drc_os)); drba->drba_cookie->drc_should_save = B_TRUE; spa_history_log_internal_ds(ds, "resume receive", tx, " "); } /* * NB: callers *MUST* call dmu_recv_stream() if dmu_recv_begin() * succeeds; otherwise we will leak the holds on the datasets. */ int dmu_recv_begin(const char *tofs, const char *tosnap, dmu_replay_record_t *drr_begin, boolean_t force, boolean_t heal, boolean_t resumable, nvlist_t *localprops, nvlist_t *hidden_args, const char *origin, dmu_recv_cookie_t *drc, zfs_file_t *fp, offset_t *voffp) { dmu_recv_begin_arg_t drba = { 0 }; int err = 0; memset(drc, 0, sizeof (dmu_recv_cookie_t)); drc->drc_drr_begin = drr_begin; drc->drc_drrb = &drr_begin->drr_u.drr_begin; drc->drc_tosnap = tosnap; drc->drc_tofs = tofs; drc->drc_force = force; drc->drc_heal = heal; drc->drc_resumable = resumable; drc->drc_cred = CRED(); drc->drc_proc = curproc; drc->drc_clone = (origin != NULL); if (drc->drc_drrb->drr_magic == BSWAP_64(DMU_BACKUP_MAGIC)) { drc->drc_byteswap = B_TRUE; (void) fletcher_4_incremental_byteswap(drr_begin, sizeof (dmu_replay_record_t), &drc->drc_cksum); byteswap_record(drr_begin); } else if (drc->drc_drrb->drr_magic == DMU_BACKUP_MAGIC) { (void) fletcher_4_incremental_native(drr_begin, sizeof (dmu_replay_record_t), &drc->drc_cksum); } else { return (SET_ERROR(EINVAL)); } drc->drc_fp = fp; drc->drc_voff = *voffp; drc->drc_featureflags = DMU_GET_FEATUREFLAGS(drc->drc_drrb->drr_versioninfo); uint32_t payloadlen = drc->drc_drr_begin->drr_payloadlen; /* * Since OpenZFS 2.0.0, we have enforced a 64MB limit in userspace * configurable via ZFS_SENDRECV_MAX_NVLIST. We enforce 256MB as a hard * upper limit. Systems with less than 1GB of RAM will see a lower * limit from `arc_all_memory() / 4`. */ if (payloadlen > (MIN((1U << 28), arc_all_memory() / 4))) return (E2BIG); if (payloadlen != 0) { void *payload = vmem_alloc(payloadlen, KM_SLEEP); /* * For compatibility with recursive send streams, we don't do * this here if the stream could be part of a package. Instead, * we'll do it in dmu_recv_stream. If we pull the next header * too early, and it's the END record, we break the `recv_skip` * logic. */ err = receive_read_payload_and_next_header(drc, payloadlen, payload); if (err != 0) { vmem_free(payload, payloadlen); return (err); } err = nvlist_unpack(payload, payloadlen, &drc->drc_begin_nvl, KM_SLEEP); vmem_free(payload, payloadlen); if (err != 0) { kmem_free(drc->drc_next_rrd, sizeof (*drc->drc_next_rrd)); return (err); } } if (drc->drc_drrb->drr_flags & DRR_FLAG_SPILL_BLOCK) drc->drc_spill = B_TRUE; drba.drba_origin = origin; drba.drba_cookie = drc; drba.drba_cred = CRED(); drba.drba_proc = curproc; if (drc->drc_featureflags & DMU_BACKUP_FEATURE_RESUMING) { err = dsl_sync_task(tofs, dmu_recv_resume_begin_check, dmu_recv_resume_begin_sync, &drba, 5, ZFS_SPACE_CHECK_NORMAL); } else { /* * For non-raw, non-incremental, non-resuming receives the * user can specify encryption parameters on the command line * with "zfs recv -o". For these receives we create a dcp and * pass it to the sync task. Creating the dcp will implicitly * remove the encryption params from the localprops nvlist, * which avoids errors when trying to set these normally * read-only properties. Any other kind of receive that * attempts to set these properties will fail as a result. */ if ((DMU_GET_FEATUREFLAGS(drc->drc_drrb->drr_versioninfo) & DMU_BACKUP_FEATURE_RAW) == 0 && origin == NULL && drc->drc_drrb->drr_fromguid == 0) { err = dsl_crypto_params_create_nvlist(DCP_CMD_NONE, localprops, hidden_args, &drba.drba_dcp); } if (err == 0) { err = dsl_sync_task(tofs, dmu_recv_begin_check, dmu_recv_begin_sync, &drba, 5, ZFS_SPACE_CHECK_NORMAL); dsl_crypto_params_free(drba.drba_dcp, !!err); } } if (err != 0) { kmem_free(drc->drc_next_rrd, sizeof (*drc->drc_next_rrd)); nvlist_free(drc->drc_begin_nvl); } return (err); } /* * Holds data need for corrective recv callback */ typedef struct cr_cb_data { uint64_t size; zbookmark_phys_t zb; spa_t *spa; } cr_cb_data_t; static void corrective_read_done(zio_t *zio) { cr_cb_data_t *data = zio->io_private; /* Corruption corrected; update error log if needed */ if (zio->io_error == 0) - spa_remove_error(data->spa, &data->zb); + spa_remove_error(data->spa, &data->zb, &zio->io_bp->blk_birth); kmem_free(data, sizeof (cr_cb_data_t)); abd_free(zio->io_abd); } /* * zio_rewrite the data pointed to by bp with the data from the rrd's abd. */ static int do_corrective_recv(struct receive_writer_arg *rwa, struct drr_write *drrw, struct receive_record_arg *rrd, blkptr_t *bp) { int err; zio_t *io; zbookmark_phys_t zb; dnode_t *dn; abd_t *abd = rrd->abd; zio_cksum_t bp_cksum = bp->blk_cksum; zio_flag_t flags = ZIO_FLAG_SPECULATIVE | ZIO_FLAG_DONT_CACHE | ZIO_FLAG_DONT_RETRY | ZIO_FLAG_CANFAIL; if (rwa->raw) flags |= ZIO_FLAG_RAW; err = dnode_hold(rwa->os, drrw->drr_object, FTAG, &dn); if (err != 0) return (err); SET_BOOKMARK(&zb, dmu_objset_id(rwa->os), drrw->drr_object, 0, dbuf_whichblock(dn, 0, drrw->drr_offset)); dnode_rele(dn, FTAG); if (!rwa->raw && DRR_WRITE_COMPRESSED(drrw)) { /* Decompress the stream data */ abd_t *dabd = abd_alloc_linear( drrw->drr_logical_size, B_FALSE); err = zio_decompress_data(drrw->drr_compressiontype, abd, abd_to_buf(dabd), abd_get_size(abd), abd_get_size(dabd), NULL); if (err != 0) { abd_free(dabd); return (err); } /* Swap in the newly decompressed data into the abd */ abd_free(abd); abd = dabd; } if (!rwa->raw && BP_GET_COMPRESS(bp) != ZIO_COMPRESS_OFF) { /* Recompress the data */ abd_t *cabd = abd_alloc_linear(BP_GET_PSIZE(bp), B_FALSE); void *buf = abd_to_buf(cabd); uint64_t csize = zio_compress_data(BP_GET_COMPRESS(bp), abd, &buf, abd_get_size(abd), rwa->os->os_complevel); abd_zero_off(cabd, csize, BP_GET_PSIZE(bp) - csize); /* Swap in newly compressed data into the abd */ abd_free(abd); abd = cabd; flags |= ZIO_FLAG_RAW_COMPRESS; } /* * The stream is not encrypted but the data on-disk is. * We need to re-encrypt the buf using the same * encryption type, salt, iv, and mac that was used to encrypt * the block previosly. */ if (!rwa->raw && BP_USES_CRYPT(bp)) { dsl_dataset_t *ds; dsl_crypto_key_t *dck = NULL; uint8_t salt[ZIO_DATA_SALT_LEN]; uint8_t iv[ZIO_DATA_IV_LEN]; uint8_t mac[ZIO_DATA_MAC_LEN]; boolean_t no_crypt = B_FALSE; dsl_pool_t *dp = dmu_objset_pool(rwa->os); abd_t *eabd = abd_alloc_linear(BP_GET_PSIZE(bp), B_FALSE); zio_crypt_decode_params_bp(bp, salt, iv); zio_crypt_decode_mac_bp(bp, mac); dsl_pool_config_enter(dp, FTAG); err = dsl_dataset_hold_flags(dp, rwa->tofs, DS_HOLD_FLAG_DECRYPT, FTAG, &ds); if (err != 0) { dsl_pool_config_exit(dp, FTAG); abd_free(eabd); return (SET_ERROR(EACCES)); } /* Look up the key from the spa's keystore */ err = spa_keystore_lookup_key(rwa->os->os_spa, zb.zb_objset, FTAG, &dck); if (err != 0) { dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); dsl_pool_config_exit(dp, FTAG); abd_free(eabd); return (SET_ERROR(EACCES)); } err = zio_do_crypt_abd(B_TRUE, &dck->dck_key, BP_GET_TYPE(bp), BP_SHOULD_BYTESWAP(bp), salt, iv, mac, abd_get_size(abd), abd, eabd, &no_crypt); spa_keystore_dsl_key_rele(rwa->os->os_spa, dck, FTAG); dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); dsl_pool_config_exit(dp, FTAG); ASSERT0(no_crypt); if (err != 0) { abd_free(eabd); return (err); } /* Swap in the newly encrypted data into the abd */ abd_free(abd); abd = eabd; /* * We want to prevent zio_rewrite() from trying to * encrypt the data again */ flags |= ZIO_FLAG_RAW_ENCRYPT; } rrd->abd = abd; io = zio_rewrite(NULL, rwa->os->os_spa, bp->blk_birth, bp, abd, BP_GET_PSIZE(bp), NULL, NULL, ZIO_PRIORITY_SYNC_WRITE, flags, &zb); ASSERT(abd_get_size(abd) == BP_GET_LSIZE(bp) || abd_get_size(abd) == BP_GET_PSIZE(bp)); /* compute new bp checksum value and make sure it matches the old one */ zio_checksum_compute(io, BP_GET_CHECKSUM(bp), abd, abd_get_size(abd)); if (!ZIO_CHECKSUM_EQUAL(bp_cksum, io->io_bp->blk_cksum)) { zio_destroy(io); if (zfs_recv_best_effort_corrective != 0) return (0); return (SET_ERROR(ECKSUM)); } /* Correct the corruption in place */ err = zio_wait(io); if (err == 0) { cr_cb_data_t *cb_data = kmem_alloc(sizeof (cr_cb_data_t), KM_SLEEP); cb_data->spa = rwa->os->os_spa; cb_data->size = drrw->drr_logical_size; cb_data->zb = zb; /* Test if healing worked by re-reading the bp */ err = zio_wait(zio_read(rwa->heal_pio, rwa->os->os_spa, bp, abd_alloc_for_io(drrw->drr_logical_size, B_FALSE), drrw->drr_logical_size, corrective_read_done, cb_data, ZIO_PRIORITY_ASYNC_READ, flags, NULL)); } if (err != 0 && zfs_recv_best_effort_corrective != 0) err = 0; return (err); } static int receive_read(dmu_recv_cookie_t *drc, int len, void *buf) { int done = 0; /* * The code doesn't rely on this (lengths being multiples of 8). See * comment in dump_bytes. */ ASSERT(len % 8 == 0 || (drc->drc_featureflags & DMU_BACKUP_FEATURE_RAW) != 0); while (done < len) { ssize_t resid = len - done; zfs_file_t *fp = drc->drc_fp; int err = zfs_file_read(fp, (char *)buf + done, len - done, &resid); if (err == 0 && resid == len - done) { /* * Note: ECKSUM or ZFS_ERR_STREAM_TRUNCATED indicates * that the receive was interrupted and can * potentially be resumed. */ err = SET_ERROR(ZFS_ERR_STREAM_TRUNCATED); } drc->drc_voff += len - done - resid; done = len - resid; if (err != 0) return (err); } drc->drc_bytes_read += len; ASSERT3U(done, ==, len); return (0); } static inline uint8_t deduce_nblkptr(dmu_object_type_t bonus_type, uint64_t bonus_size) { if (bonus_type == DMU_OT_SA) { return (1); } else { return (1 + ((DN_OLD_MAX_BONUSLEN - MIN(DN_OLD_MAX_BONUSLEN, bonus_size)) >> SPA_BLKPTRSHIFT)); } } static void save_resume_state(struct receive_writer_arg *rwa, uint64_t object, uint64_t offset, dmu_tx_t *tx) { int txgoff = dmu_tx_get_txg(tx) & TXG_MASK; if (!rwa->resumable) return; /* * We use ds_resume_bytes[] != 0 to indicate that we need to * update this on disk, so it must not be 0. */ ASSERT(rwa->bytes_read != 0); /* * We only resume from write records, which have a valid * (non-meta-dnode) object number. */ ASSERT(object != 0); /* * For resuming to work correctly, we must receive records in order, * sorted by object,offset. This is checked by the callers, but * assert it here for good measure. */ ASSERT3U(object, >=, rwa->os->os_dsl_dataset->ds_resume_object[txgoff]); ASSERT(object != rwa->os->os_dsl_dataset->ds_resume_object[txgoff] || offset >= rwa->os->os_dsl_dataset->ds_resume_offset[txgoff]); ASSERT3U(rwa->bytes_read, >=, rwa->os->os_dsl_dataset->ds_resume_bytes[txgoff]); rwa->os->os_dsl_dataset->ds_resume_object[txgoff] = object; rwa->os->os_dsl_dataset->ds_resume_offset[txgoff] = offset; rwa->os->os_dsl_dataset->ds_resume_bytes[txgoff] = rwa->bytes_read; } static int receive_object_is_same_generation(objset_t *os, uint64_t object, dmu_object_type_t old_bonus_type, dmu_object_type_t new_bonus_type, const void *new_bonus, boolean_t *samegenp) { zfs_file_info_t zoi; int err; dmu_buf_t *old_bonus_dbuf; err = dmu_bonus_hold(os, object, FTAG, &old_bonus_dbuf); if (err != 0) return (err); err = dmu_get_file_info(os, old_bonus_type, old_bonus_dbuf->db_data, &zoi); dmu_buf_rele(old_bonus_dbuf, FTAG); if (err != 0) return (err); uint64_t old_gen = zoi.zfi_generation; err = dmu_get_file_info(os, new_bonus_type, new_bonus, &zoi); if (err != 0) return (err); uint64_t new_gen = zoi.zfi_generation; *samegenp = (old_gen == new_gen); return (0); } static int receive_handle_existing_object(const struct receive_writer_arg *rwa, const struct drr_object *drro, const dmu_object_info_t *doi, const void *bonus_data, uint64_t *object_to_hold, uint32_t *new_blksz) { uint32_t indblksz = drro->drr_indblkshift ? 1ULL << drro->drr_indblkshift : 0; int nblkptr = deduce_nblkptr(drro->drr_bonustype, drro->drr_bonuslen); uint8_t dn_slots = drro->drr_dn_slots != 0 ? drro->drr_dn_slots : DNODE_MIN_SLOTS; boolean_t do_free_range = B_FALSE; int err; *object_to_hold = drro->drr_object; /* nblkptr should be bounded by the bonus size and type */ if (rwa->raw && nblkptr != drro->drr_nblkptr) return (SET_ERROR(EINVAL)); /* * After the previous send stream, the sending system may * have freed this object, and then happened to re-allocate * this object number in a later txg. In this case, we are * receiving a different logical file, and the block size may * appear to be different. i.e. we may have a different * block size for this object than what the send stream says. * In this case we need to remove the object's contents, * so that its structure can be changed and then its contents * entirely replaced by subsequent WRITE records. * * If this is a -L (--large-block) incremental stream, and * the previous stream was not -L, the block size may appear * to increase. i.e. we may have a smaller block size for * this object than what the send stream says. In this case * we need to keep the object's contents and block size * intact, so that we don't lose parts of the object's * contents that are not changed by this incremental send * stream. * * We can distinguish between the two above cases by using * the ZPL's generation number (see * receive_object_is_same_generation()). However, we only * want to rely on the generation number when absolutely * necessary, because with raw receives, the generation is * encrypted. We also want to minimize dependence on the * ZPL, so that other types of datasets can also be received * (e.g. ZVOLs, although note that ZVOLS currently do not * reallocate their objects or change their structure). * Therefore, we check a number of different cases where we * know it is safe to discard the object's contents, before * using the ZPL's generation number to make the above * distinction. */ if (drro->drr_blksz != doi->doi_data_block_size) { if (rwa->raw) { /* * RAW streams always have large blocks, so * we are sure that the data is not needed * due to changing --large-block to be on. * Which is fortunate since the bonus buffer * (which contains the ZPL generation) is * encrypted, and the key might not be * loaded. */ do_free_range = B_TRUE; } else if (rwa->full) { /* * This is a full send stream, so it always * replaces what we have. Even if the * generation numbers happen to match, this * can not actually be the same logical file. * This is relevant when receiving a full * send as a clone. */ do_free_range = B_TRUE; } else if (drro->drr_type != DMU_OT_PLAIN_FILE_CONTENTS || doi->doi_type != DMU_OT_PLAIN_FILE_CONTENTS) { /* * PLAIN_FILE_CONTENTS are the only type of * objects that have ever been stored with * large blocks, so we don't need the special * logic below. ZAP blocks can shrink (when * there's only one block), so we don't want * to hit the error below about block size * only increasing. */ do_free_range = B_TRUE; } else if (doi->doi_max_offset <= doi->doi_data_block_size) { /* * There is only one block. We can free it, * because its contents will be replaced by a * WRITE record. This can not be the no-L -> * -L case, because the no-L case would have * resulted in multiple blocks. If we * supported -L -> no-L, it would not be safe * to free the file's contents. Fortunately, * that is not allowed (see * recv_check_large_blocks()). */ do_free_range = B_TRUE; } else { boolean_t is_same_gen; err = receive_object_is_same_generation(rwa->os, drro->drr_object, doi->doi_bonus_type, drro->drr_bonustype, bonus_data, &is_same_gen); if (err != 0) return (SET_ERROR(EINVAL)); if (is_same_gen) { /* * This is the same logical file, and * the block size must be increasing. * It could only decrease if * --large-block was changed to be * off, which is checked in * recv_check_large_blocks(). */ if (drro->drr_blksz <= doi->doi_data_block_size) return (SET_ERROR(EINVAL)); /* * We keep the existing blocksize and * contents. */ *new_blksz = doi->doi_data_block_size; } else { do_free_range = B_TRUE; } } } /* nblkptr can only decrease if the object was reallocated */ if (nblkptr < doi->doi_nblkptr) do_free_range = B_TRUE; /* number of slots can only change on reallocation */ if (dn_slots != doi->doi_dnodesize >> DNODE_SHIFT) do_free_range = B_TRUE; /* * For raw sends we also check a few other fields to * ensure we are preserving the objset structure exactly * as it was on the receive side: * - A changed indirect block size * - A smaller nlevels */ if (rwa->raw) { if (indblksz != doi->doi_metadata_block_size) do_free_range = B_TRUE; if (drro->drr_nlevels < doi->doi_indirection) do_free_range = B_TRUE; } if (do_free_range) { err = dmu_free_long_range(rwa->os, drro->drr_object, 0, DMU_OBJECT_END); if (err != 0) return (SET_ERROR(EINVAL)); } /* * The dmu does not currently support decreasing nlevels * or changing the number of dnode slots on an object. For * non-raw sends, this does not matter and the new object * can just use the previous one's nlevels. For raw sends, * however, the structure of the received dnode (including * nlevels and dnode slots) must match that of the send * side. Therefore, instead of using dmu_object_reclaim(), * we must free the object completely and call * dmu_object_claim_dnsize() instead. */ if ((rwa->raw && drro->drr_nlevels < doi->doi_indirection) || dn_slots != doi->doi_dnodesize >> DNODE_SHIFT) { err = dmu_free_long_object(rwa->os, drro->drr_object); if (err != 0) return (SET_ERROR(EINVAL)); txg_wait_synced(dmu_objset_pool(rwa->os), 0); *object_to_hold = DMU_NEW_OBJECT; } /* * For raw receives, free everything beyond the new incoming * maxblkid. Normally this would be done with a DRR_FREE * record that would come after this DRR_OBJECT record is * processed. However, for raw receives we manually set the * maxblkid from the drr_maxblkid and so we must first free * everything above that blkid to ensure the DMU is always * consistent with itself. We will never free the first block * of the object here because a maxblkid of 0 could indicate * an object with a single block or one with no blocks. This * free may be skipped when dmu_free_long_range() was called * above since it covers the entire object's contents. */ if (rwa->raw && *object_to_hold != DMU_NEW_OBJECT && !do_free_range) { err = dmu_free_long_range(rwa->os, drro->drr_object, (drro->drr_maxblkid + 1) * doi->doi_data_block_size, DMU_OBJECT_END); if (err != 0) return (SET_ERROR(EINVAL)); } return (0); } noinline static int receive_object(struct receive_writer_arg *rwa, struct drr_object *drro, void *data) { dmu_object_info_t doi; dmu_tx_t *tx; int err; uint32_t new_blksz = drro->drr_blksz; uint8_t dn_slots = drro->drr_dn_slots != 0 ? drro->drr_dn_slots : DNODE_MIN_SLOTS; if (drro->drr_type == DMU_OT_NONE || !DMU_OT_IS_VALID(drro->drr_type) || !DMU_OT_IS_VALID(drro->drr_bonustype) || drro->drr_checksumtype >= ZIO_CHECKSUM_FUNCTIONS || drro->drr_compress >= ZIO_COMPRESS_FUNCTIONS || P2PHASE(drro->drr_blksz, SPA_MINBLOCKSIZE) || drro->drr_blksz < SPA_MINBLOCKSIZE || drro->drr_blksz > spa_maxblocksize(dmu_objset_spa(rwa->os)) || drro->drr_bonuslen > DN_BONUS_SIZE(spa_maxdnodesize(dmu_objset_spa(rwa->os))) || dn_slots > (spa_maxdnodesize(dmu_objset_spa(rwa->os)) >> DNODE_SHIFT)) { return (SET_ERROR(EINVAL)); } if (rwa->raw) { /* * We should have received a DRR_OBJECT_RANGE record * containing this block and stored it in rwa. */ if (drro->drr_object < rwa->or_firstobj || drro->drr_object >= rwa->or_firstobj + rwa->or_numslots || drro->drr_raw_bonuslen < drro->drr_bonuslen || drro->drr_indblkshift > SPA_MAXBLOCKSHIFT || drro->drr_nlevels > DN_MAX_LEVELS || drro->drr_nblkptr > DN_MAX_NBLKPTR || DN_SLOTS_TO_BONUSLEN(dn_slots) < drro->drr_raw_bonuslen) return (SET_ERROR(EINVAL)); } else { /* * The DRR_OBJECT_SPILL flag is valid when the DRR_BEGIN * record indicates this by setting DRR_FLAG_SPILL_BLOCK. */ if (((drro->drr_flags & ~(DRR_OBJECT_SPILL))) || (!rwa->spill && DRR_OBJECT_HAS_SPILL(drro->drr_flags))) { return (SET_ERROR(EINVAL)); } if (drro->drr_raw_bonuslen != 0 || drro->drr_nblkptr != 0 || drro->drr_indblkshift != 0 || drro->drr_nlevels != 0) { return (SET_ERROR(EINVAL)); } } err = dmu_object_info(rwa->os, drro->drr_object, &doi); if (err != 0 && err != ENOENT && err != EEXIST) return (SET_ERROR(EINVAL)); if (drro->drr_object > rwa->max_object) rwa->max_object = drro->drr_object; /* * If we are losing blkptrs or changing the block size this must * be a new file instance. We must clear out the previous file * contents before we can change this type of metadata in the dnode. * Raw receives will also check that the indirect structure of the * dnode hasn't changed. */ uint64_t object_to_hold; if (err == 0) { err = receive_handle_existing_object(rwa, drro, &doi, data, &object_to_hold, &new_blksz); if (err != 0) return (err); } else if (err == EEXIST) { /* * The object requested is currently an interior slot of a * multi-slot dnode. This will be resolved when the next txg * is synced out, since the send stream will have told us * to free this slot when we freed the associated dnode * earlier in the stream. */ txg_wait_synced(dmu_objset_pool(rwa->os), 0); if (dmu_object_info(rwa->os, drro->drr_object, NULL) != ENOENT) return (SET_ERROR(EINVAL)); /* object was freed and we are about to allocate a new one */ object_to_hold = DMU_NEW_OBJECT; } else { /* * If the only record in this range so far was DRR_FREEOBJECTS * with at least one actually freed object, it's possible that * the block will now be converted to a hole. We need to wait * for the txg to sync to prevent races. */ if (rwa->or_need_sync == ORNS_YES) txg_wait_synced(dmu_objset_pool(rwa->os), 0); /* object is free and we are about to allocate a new one */ object_to_hold = DMU_NEW_OBJECT; } /* Only relevant for the first object in the range */ rwa->or_need_sync = ORNS_NO; /* * If this is a multi-slot dnode there is a chance that this * object will expand into a slot that is already used by * another object from the previous snapshot. We must free * these objects before we attempt to allocate the new dnode. */ if (dn_slots > 1) { boolean_t need_sync = B_FALSE; for (uint64_t slot = drro->drr_object + 1; slot < drro->drr_object + dn_slots; slot++) { dmu_object_info_t slot_doi; err = dmu_object_info(rwa->os, slot, &slot_doi); if (err == ENOENT || err == EEXIST) continue; else if (err != 0) return (err); err = dmu_free_long_object(rwa->os, slot); if (err != 0) return (err); need_sync = B_TRUE; } if (need_sync) txg_wait_synced(dmu_objset_pool(rwa->os), 0); } tx = dmu_tx_create(rwa->os); dmu_tx_hold_bonus(tx, object_to_hold); dmu_tx_hold_write(tx, object_to_hold, 0, 0); err = dmu_tx_assign(tx, TXG_WAIT); if (err != 0) { dmu_tx_abort(tx); return (err); } if (object_to_hold == DMU_NEW_OBJECT) { /* Currently free, wants to be allocated */ err = dmu_object_claim_dnsize(rwa->os, drro->drr_object, drro->drr_type, new_blksz, drro->drr_bonustype, drro->drr_bonuslen, dn_slots << DNODE_SHIFT, tx); } else if (drro->drr_type != doi.doi_type || new_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_dnsize(rwa->os, drro->drr_object, drro->drr_type, new_blksz, drro->drr_bonustype, drro->drr_bonuslen, dn_slots << DNODE_SHIFT, rwa->spill ? DRR_OBJECT_HAS_SPILL(drro->drr_flags) : B_FALSE, tx); } else if (rwa->spill && !DRR_OBJECT_HAS_SPILL(drro->drr_flags)) { /* * Currently allocated, the existing version of this object * may reference a spill block that is no longer allocated * at the source and needs to be freed. */ err = dmu_object_rm_spill(rwa->os, drro->drr_object, tx); } if (err != 0) { dmu_tx_commit(tx); return (SET_ERROR(EINVAL)); } if (rwa->or_crypt_params_present) { /* * Set the crypt params for the buffer associated with this * range of dnodes. This causes the blkptr_t to have the * same crypt params (byteorder, salt, iv, mac) as on the * sending side. * * Since we are committing this tx now, it is possible for * the dnode block to end up on-disk with the incorrect MAC, * if subsequent objects in this block are received in a * different txg. However, since the dataset is marked as * inconsistent, no code paths will do a non-raw read (or * decrypt the block / verify the MAC). The receive code and * scrub code can safely do raw reads and verify the * checksum. They don't need to verify the MAC. */ dmu_buf_t *db = NULL; uint64_t offset = rwa->or_firstobj * DNODE_MIN_SIZE; err = dmu_buf_hold_by_dnode(DMU_META_DNODE(rwa->os), offset, FTAG, &db, DMU_READ_PREFETCH | DMU_READ_NO_DECRYPT); if (err != 0) { dmu_tx_commit(tx); return (SET_ERROR(EINVAL)); } dmu_buf_set_crypt_params(db, rwa->or_byteorder, rwa->or_salt, rwa->or_iv, rwa->or_mac, tx); dmu_buf_rele(db, FTAG); rwa->or_crypt_params_present = B_FALSE; } dmu_object_set_checksum(rwa->os, drro->drr_object, drro->drr_checksumtype, tx); dmu_object_set_compress(rwa->os, drro->drr_object, drro->drr_compress, tx); /* handle more restrictive dnode structuring for raw recvs */ if (rwa->raw) { /* * Set the indirect block size, block shift, nlevels. * This will not fail because we ensured all of the * blocks were freed earlier if this is a new object. * For non-new objects block size and indirect block * shift cannot change and nlevels can only increase. */ ASSERT3U(new_blksz, ==, drro->drr_blksz); VERIFY0(dmu_object_set_blocksize(rwa->os, drro->drr_object, drro->drr_blksz, drro->drr_indblkshift, tx)); VERIFY0(dmu_object_set_nlevels(rwa->os, drro->drr_object, drro->drr_nlevels, tx)); /* * Set the maxblkid. This will always succeed because * we freed all blocks beyond the new maxblkid above. */ VERIFY0(dmu_object_set_maxblkid(rwa->os, drro->drr_object, drro->drr_maxblkid, tx)); } if (data != NULL) { dmu_buf_t *db; dnode_t *dn; uint32_t flags = DMU_READ_NO_PREFETCH; if (rwa->raw) flags |= DMU_READ_NO_DECRYPT; VERIFY0(dnode_hold(rwa->os, drro->drr_object, FTAG, &dn)); VERIFY0(dmu_bonus_hold_by_dnode(dn, FTAG, &db, flags)); dmu_buf_will_dirty(db, tx); ASSERT3U(db->db_size, >=, drro->drr_bonuslen); memcpy(db->db_data, data, DRR_OBJECT_PAYLOAD_SIZE(drro)); /* * Raw bonus buffers have their byteorder determined by the * DRR_OBJECT_RANGE record. */ if (rwa->byteswap && !rwa->raw) { dmu_object_byteswap_t byteswap = DMU_OT_BYTESWAP(drro->drr_bonustype); dmu_ot_byteswap[byteswap].ob_func(db->db_data, DRR_OBJECT_PAYLOAD_SIZE(drro)); } dmu_buf_rele(db, FTAG); dnode_rele(dn, FTAG); } dmu_tx_commit(tx); return (0); } noinline static int receive_freeobjects(struct receive_writer_arg *rwa, struct drr_freeobjects *drrfo) { uint64_t obj; int next_err = 0; if (drrfo->drr_firstobj + drrfo->drr_numobjs < drrfo->drr_firstobj) return (SET_ERROR(EINVAL)); for (obj = drrfo->drr_firstobj == 0 ? 1 : drrfo->drr_firstobj; obj < drrfo->drr_firstobj + drrfo->drr_numobjs && obj < DN_MAX_OBJECT && next_err == 0; next_err = dmu_object_next(rwa->os, &obj, FALSE, 0)) { dmu_object_info_t doi; int err; err = dmu_object_info(rwa->os, obj, &doi); if (err == ENOENT) continue; else if (err != 0) return (err); err = dmu_free_long_object(rwa->os, obj); if (err != 0) return (err); if (rwa->or_need_sync == ORNS_MAYBE) rwa->or_need_sync = ORNS_YES; } if (next_err != ESRCH) return (next_err); return (0); } /* * Note: if this fails, the caller will clean up any records left on the * rwa->write_batch list. */ static int flush_write_batch_impl(struct receive_writer_arg *rwa) { dnode_t *dn; int err; if (dnode_hold(rwa->os, rwa->last_object, FTAG, &dn) != 0) return (SET_ERROR(EINVAL)); struct receive_record_arg *last_rrd = list_tail(&rwa->write_batch); struct drr_write *last_drrw = &last_rrd->header.drr_u.drr_write; struct receive_record_arg *first_rrd = list_head(&rwa->write_batch); struct drr_write *first_drrw = &first_rrd->header.drr_u.drr_write; ASSERT3U(rwa->last_object, ==, last_drrw->drr_object); ASSERT3U(rwa->last_offset, ==, last_drrw->drr_offset); dmu_tx_t *tx = dmu_tx_create(rwa->os); dmu_tx_hold_write_by_dnode(tx, dn, first_drrw->drr_offset, last_drrw->drr_offset - first_drrw->drr_offset + last_drrw->drr_logical_size); err = dmu_tx_assign(tx, TXG_WAIT); if (err != 0) { dmu_tx_abort(tx); dnode_rele(dn, FTAG); return (err); } struct receive_record_arg *rrd; while ((rrd = list_head(&rwa->write_batch)) != NULL) { struct drr_write *drrw = &rrd->header.drr_u.drr_write; abd_t *abd = rrd->abd; ASSERT3U(drrw->drr_object, ==, rwa->last_object); if (drrw->drr_logical_size != dn->dn_datablksz) { /* * The WRITE record is larger than the object's block * size. We must be receiving an incremental * large-block stream into a dataset that previously did * a non-large-block receive. Lightweight writes must * be exactly one block, so we need to decompress the * data (if compressed) and do a normal dmu_write(). */ ASSERT3U(drrw->drr_logical_size, >, dn->dn_datablksz); if (DRR_WRITE_COMPRESSED(drrw)) { abd_t *decomp_abd = abd_alloc_linear(drrw->drr_logical_size, B_FALSE); err = zio_decompress_data( drrw->drr_compressiontype, abd, abd_to_buf(decomp_abd), abd_get_size(abd), abd_get_size(decomp_abd), NULL); if (err == 0) { dmu_write_by_dnode(dn, drrw->drr_offset, drrw->drr_logical_size, abd_to_buf(decomp_abd), tx); } abd_free(decomp_abd); } else { dmu_write_by_dnode(dn, drrw->drr_offset, drrw->drr_logical_size, abd_to_buf(abd), tx); } if (err == 0) abd_free(abd); } else { zio_prop_t zp = {0}; dmu_write_policy(rwa->os, dn, 0, 0, &zp); zio_flag_t zio_flags = 0; if (rwa->raw) { zp.zp_encrypt = B_TRUE; zp.zp_compress = drrw->drr_compressiontype; zp.zp_byteorder = ZFS_HOST_BYTEORDER ^ !!DRR_IS_RAW_BYTESWAPPED(drrw->drr_flags) ^ rwa->byteswap; memcpy(zp.zp_salt, drrw->drr_salt, ZIO_DATA_SALT_LEN); memcpy(zp.zp_iv, drrw->drr_iv, ZIO_DATA_IV_LEN); memcpy(zp.zp_mac, drrw->drr_mac, ZIO_DATA_MAC_LEN); if (DMU_OT_IS_ENCRYPTED(zp.zp_type)) { zp.zp_nopwrite = B_FALSE; zp.zp_copies = MIN(zp.zp_copies, SPA_DVAS_PER_BP - 1); } zio_flags |= ZIO_FLAG_RAW; } else if (DRR_WRITE_COMPRESSED(drrw)) { ASSERT3U(drrw->drr_compressed_size, >, 0); ASSERT3U(drrw->drr_logical_size, >=, drrw->drr_compressed_size); zp.zp_compress = drrw->drr_compressiontype; zio_flags |= ZIO_FLAG_RAW_COMPRESS; } else if (rwa->byteswap) { /* * Note: compressed blocks never need to be * byteswapped, because WRITE records for * metadata blocks are never compressed. The * exception is raw streams, which are written * in the original byteorder, and the byteorder * bit is preserved in the BP by setting * zp_byteorder above. */ dmu_object_byteswap_t byteswap = DMU_OT_BYTESWAP(drrw->drr_type); dmu_ot_byteswap[byteswap].ob_func( abd_to_buf(abd), DRR_WRITE_PAYLOAD_SIZE(drrw)); } /* * Since this data can't be read until the receive * completes, we can do a "lightweight" write for * improved performance. */ err = dmu_lightweight_write_by_dnode(dn, drrw->drr_offset, abd, &zp, zio_flags, tx); } if (err != 0) { /* * This rrd is left on the list, so the caller will * free it (and the abd). */ break; } /* * Note: If the receive fails, we want the resume stream to * start with the same record that we last successfully * received (as opposed to the next record), so that we can * verify that we are resuming from the correct location. */ save_resume_state(rwa, drrw->drr_object, drrw->drr_offset, tx); list_remove(&rwa->write_batch, rrd); kmem_free(rrd, sizeof (*rrd)); } dmu_tx_commit(tx); dnode_rele(dn, FTAG); return (err); } noinline static int flush_write_batch(struct receive_writer_arg *rwa) { if (list_is_empty(&rwa->write_batch)) return (0); int err = rwa->err; if (err == 0) err = flush_write_batch_impl(rwa); if (err != 0) { struct receive_record_arg *rrd; while ((rrd = list_remove_head(&rwa->write_batch)) != NULL) { abd_free(rrd->abd); kmem_free(rrd, sizeof (*rrd)); } } ASSERT(list_is_empty(&rwa->write_batch)); return (err); } noinline static int receive_process_write_record(struct receive_writer_arg *rwa, struct receive_record_arg *rrd) { int err = 0; ASSERT3U(rrd->header.drr_type, ==, DRR_WRITE); struct drr_write *drrw = &rrd->header.drr_u.drr_write; if (drrw->drr_offset + drrw->drr_logical_size < drrw->drr_offset || !DMU_OT_IS_VALID(drrw->drr_type)) return (SET_ERROR(EINVAL)); if (rwa->heal) { blkptr_t *bp; dmu_buf_t *dbp; dnode_t *dn; int flags = DB_RF_CANFAIL; if (rwa->raw) flags |= DB_RF_NO_DECRYPT; if (rwa->byteswap) { dmu_object_byteswap_t byteswap = DMU_OT_BYTESWAP(drrw->drr_type); dmu_ot_byteswap[byteswap].ob_func(abd_to_buf(rrd->abd), DRR_WRITE_PAYLOAD_SIZE(drrw)); } err = dmu_buf_hold_noread(rwa->os, drrw->drr_object, drrw->drr_offset, FTAG, &dbp); if (err != 0) return (err); /* Try to read the object to see if it needs healing */ err = dbuf_read((dmu_buf_impl_t *)dbp, NULL, flags); /* * We only try to heal when dbuf_read() returns a ECKSUMs. * Other errors (even EIO) get returned to caller. * EIO indicates that the device is not present/accessible, * so writing to it will likely fail. * If the block is healthy, we don't want to overwrite it * unnecessarily. */ if (err != ECKSUM) { dmu_buf_rele(dbp, FTAG); return (err); } dn = dmu_buf_dnode_enter(dbp); /* Make sure the on-disk block and recv record sizes match */ if (drrw->drr_logical_size != dn->dn_datablkszsec << SPA_MINBLOCKSHIFT) { err = ENOTSUP; dmu_buf_dnode_exit(dbp); dmu_buf_rele(dbp, FTAG); return (err); } /* Get the block pointer for the corrupted block */ bp = dmu_buf_get_blkptr(dbp); err = do_corrective_recv(rwa, drrw, rrd, bp); dmu_buf_dnode_exit(dbp); dmu_buf_rele(dbp, FTAG); return (err); } /* * For resuming to work, records must be in increasing order * by (object, offset). */ if (drrw->drr_object < rwa->last_object || (drrw->drr_object == rwa->last_object && drrw->drr_offset < rwa->last_offset)) { return (SET_ERROR(EINVAL)); } struct receive_record_arg *first_rrd = list_head(&rwa->write_batch); struct drr_write *first_drrw = &first_rrd->header.drr_u.drr_write; uint64_t batch_size = MIN(zfs_recv_write_batch_size, DMU_MAX_ACCESS / 2); if (first_rrd != NULL && (drrw->drr_object != first_drrw->drr_object || drrw->drr_offset >= first_drrw->drr_offset + batch_size)) { err = flush_write_batch(rwa); if (err != 0) return (err); } rwa->last_object = drrw->drr_object; rwa->last_offset = drrw->drr_offset; if (rwa->last_object > rwa->max_object) rwa->max_object = rwa->last_object; list_insert_tail(&rwa->write_batch, rrd); /* * Return EAGAIN to indicate that we will use this rrd again, * so the caller should not free it */ return (EAGAIN); } static int receive_write_embedded(struct receive_writer_arg *rwa, struct drr_write_embedded *drrwe, void *data) { dmu_tx_t *tx; int err; if (drrwe->drr_offset + drrwe->drr_length < drrwe->drr_offset) return (SET_ERROR(EINVAL)); if (drrwe->drr_psize > BPE_PAYLOAD_SIZE) return (SET_ERROR(EINVAL)); if (drrwe->drr_etype >= NUM_BP_EMBEDDED_TYPES) return (SET_ERROR(EINVAL)); if (drrwe->drr_compression >= ZIO_COMPRESS_FUNCTIONS) return (SET_ERROR(EINVAL)); if (rwa->raw) return (SET_ERROR(EINVAL)); if (drrwe->drr_object > rwa->max_object) rwa->max_object = drrwe->drr_object; tx = dmu_tx_create(rwa->os); dmu_tx_hold_write(tx, drrwe->drr_object, drrwe->drr_offset, drrwe->drr_length); err = dmu_tx_assign(tx, TXG_WAIT); if (err != 0) { dmu_tx_abort(tx); return (err); } dmu_write_embedded(rwa->os, drrwe->drr_object, drrwe->drr_offset, data, drrwe->drr_etype, drrwe->drr_compression, drrwe->drr_lsize, drrwe->drr_psize, rwa->byteswap ^ ZFS_HOST_BYTEORDER, tx); /* See comment in restore_write. */ save_resume_state(rwa, drrwe->drr_object, drrwe->drr_offset, tx); dmu_tx_commit(tx); return (0); } static int receive_spill(struct receive_writer_arg *rwa, struct drr_spill *drrs, abd_t *abd) { dmu_buf_t *db, *db_spill; int err; if (drrs->drr_length < SPA_MINBLOCKSIZE || drrs->drr_length > spa_maxblocksize(dmu_objset_spa(rwa->os))) return (SET_ERROR(EINVAL)); /* * This is an unmodified spill block which was added to the stream * to resolve an issue with incorrectly removing spill blocks. It * should be ignored by current versions of the code which support * the DRR_FLAG_SPILL_BLOCK flag. */ if (rwa->spill && DRR_SPILL_IS_UNMODIFIED(drrs->drr_flags)) { abd_free(abd); return (0); } if (rwa->raw) { if (!DMU_OT_IS_VALID(drrs->drr_type) || drrs->drr_compressiontype >= ZIO_COMPRESS_FUNCTIONS || drrs->drr_compressed_size == 0) return (SET_ERROR(EINVAL)); } if (dmu_object_info(rwa->os, drrs->drr_object, NULL) != 0) return (SET_ERROR(EINVAL)); if (drrs->drr_object > rwa->max_object) rwa->max_object = drrs->drr_object; VERIFY0(dmu_bonus_hold(rwa->os, drrs->drr_object, FTAG, &db)); if ((err = dmu_spill_hold_by_bonus(db, DMU_READ_NO_DECRYPT, FTAG, &db_spill)) != 0) { dmu_buf_rele(db, FTAG); return (err); } dmu_tx_t *tx = dmu_tx_create(rwa->os); dmu_tx_hold_spill(tx, db->db_object); err = dmu_tx_assign(tx, TXG_WAIT); if (err != 0) { dmu_buf_rele(db, FTAG); dmu_buf_rele(db_spill, FTAG); dmu_tx_abort(tx); return (err); } /* * Spill blocks may both grow and shrink. When a change in size * occurs any existing dbuf must be updated to match the logical * size of the provided arc_buf_t. */ if (db_spill->db_size != drrs->drr_length) { dmu_buf_will_fill(db_spill, tx); VERIFY0(dbuf_spill_set_blksz(db_spill, drrs->drr_length, tx)); } arc_buf_t *abuf; if (rwa->raw) { boolean_t byteorder = ZFS_HOST_BYTEORDER ^ !!DRR_IS_RAW_BYTESWAPPED(drrs->drr_flags) ^ rwa->byteswap; abuf = arc_loan_raw_buf(dmu_objset_spa(rwa->os), drrs->drr_object, byteorder, drrs->drr_salt, drrs->drr_iv, drrs->drr_mac, drrs->drr_type, drrs->drr_compressed_size, drrs->drr_length, drrs->drr_compressiontype, 0); } else { abuf = arc_loan_buf(dmu_objset_spa(rwa->os), DMU_OT_IS_METADATA(drrs->drr_type), drrs->drr_length); if (rwa->byteswap) { dmu_object_byteswap_t byteswap = DMU_OT_BYTESWAP(drrs->drr_type); dmu_ot_byteswap[byteswap].ob_func(abd_to_buf(abd), DRR_SPILL_PAYLOAD_SIZE(drrs)); } } memcpy(abuf->b_data, abd_to_buf(abd), DRR_SPILL_PAYLOAD_SIZE(drrs)); abd_free(abd); dbuf_assign_arcbuf((dmu_buf_impl_t *)db_spill, abuf, tx); dmu_buf_rele(db, FTAG); dmu_buf_rele(db_spill, FTAG); dmu_tx_commit(tx); return (0); } noinline static int receive_free(struct receive_writer_arg *rwa, struct drr_free *drrf) { int err; if (drrf->drr_length != -1ULL && drrf->drr_offset + drrf->drr_length < drrf->drr_offset) return (SET_ERROR(EINVAL)); if (dmu_object_info(rwa->os, drrf->drr_object, NULL) != 0) return (SET_ERROR(EINVAL)); if (drrf->drr_object > rwa->max_object) rwa->max_object = drrf->drr_object; err = dmu_free_long_range(rwa->os, drrf->drr_object, drrf->drr_offset, drrf->drr_length); return (err); } static int receive_object_range(struct receive_writer_arg *rwa, struct drr_object_range *drror) { /* * By default, we assume this block is in our native format * (ZFS_HOST_BYTEORDER). We then take into account whether * the send stream is byteswapped (rwa->byteswap). Finally, * we need to byteswap again if this particular block was * in non-native format on the send side. */ boolean_t byteorder = ZFS_HOST_BYTEORDER ^ rwa->byteswap ^ !!DRR_IS_RAW_BYTESWAPPED(drror->drr_flags); /* * Since dnode block sizes are constant, we should not need to worry * about making sure that the dnode block size is the same on the * sending and receiving sides for the time being. For non-raw sends, * this does not matter (and in fact we do not send a DRR_OBJECT_RANGE * record at all). Raw sends require this record type because the * encryption parameters are used to protect an entire block of bonus * buffers. If the size of dnode blocks ever becomes variable, * handling will need to be added to ensure that dnode block sizes * match on the sending and receiving side. */ if (drror->drr_numslots != DNODES_PER_BLOCK || P2PHASE(drror->drr_firstobj, DNODES_PER_BLOCK) != 0 || !rwa->raw) return (SET_ERROR(EINVAL)); if (drror->drr_firstobj > rwa->max_object) rwa->max_object = drror->drr_firstobj; /* * The DRR_OBJECT_RANGE handling must be deferred to receive_object() * so that the block of dnodes is not written out when it's empty, * and converted to a HOLE BP. */ rwa->or_crypt_params_present = B_TRUE; rwa->or_firstobj = drror->drr_firstobj; rwa->or_numslots = drror->drr_numslots; memcpy(rwa->or_salt, drror->drr_salt, ZIO_DATA_SALT_LEN); memcpy(rwa->or_iv, drror->drr_iv, ZIO_DATA_IV_LEN); memcpy(rwa->or_mac, drror->drr_mac, ZIO_DATA_MAC_LEN); rwa->or_byteorder = byteorder; rwa->or_need_sync = ORNS_MAYBE; return (0); } /* * Until we have the ability to redact large ranges of data efficiently, we * process these records as frees. */ noinline static int receive_redact(struct receive_writer_arg *rwa, struct drr_redact *drrr) { struct drr_free drrf = {0}; drrf.drr_length = drrr->drr_length; drrf.drr_object = drrr->drr_object; drrf.drr_offset = drrr->drr_offset; drrf.drr_toguid = drrr->drr_toguid; return (receive_free(rwa, &drrf)); } /* used to destroy the drc_ds on error */ static void dmu_recv_cleanup_ds(dmu_recv_cookie_t *drc) { dsl_dataset_t *ds = drc->drc_ds; ds_hold_flags_t dsflags; dsflags = (drc->drc_raw) ? DS_HOLD_FLAG_NONE : DS_HOLD_FLAG_DECRYPT; /* * Wait for the txg sync before cleaning up the receive. For * resumable receives, this ensures that our resume state has * been written out to disk. For raw receives, this ensures * that the user accounting code will not attempt to do anything * after we stopped receiving the dataset. */ txg_wait_synced(ds->ds_dir->dd_pool, 0); ds->ds_objset->os_raw_receive = B_FALSE; rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG); if (drc->drc_resumable && drc->drc_should_save && !BP_IS_HOLE(dsl_dataset_get_blkptr(ds))) { rrw_exit(&ds->ds_bp_rwlock, FTAG); dsl_dataset_disown(ds, dsflags, dmu_recv_tag); } else { char name[ZFS_MAX_DATASET_NAME_LEN]; rrw_exit(&ds->ds_bp_rwlock, FTAG); dsl_dataset_name(ds, name); dsl_dataset_disown(ds, dsflags, dmu_recv_tag); if (!drc->drc_heal) (void) dsl_destroy_head(name); } } static void receive_cksum(dmu_recv_cookie_t *drc, int len, void *buf) { if (drc->drc_byteswap) { (void) fletcher_4_incremental_byteswap(buf, len, &drc->drc_cksum); } else { (void) fletcher_4_incremental_native(buf, len, &drc->drc_cksum); } } /* * Read the payload into a buffer of size len, and update the current record's * payload field. * Allocate drc->drc_next_rrd and read the next record's header into * drc->drc_next_rrd->header. * Verify checksum of payload and next record. */ static int receive_read_payload_and_next_header(dmu_recv_cookie_t *drc, int len, void *buf) { int err; if (len != 0) { ASSERT3U(len, <=, SPA_MAXBLOCKSIZE); err = receive_read(drc, len, buf); if (err != 0) return (err); receive_cksum(drc, len, buf); /* note: rrd is NULL when reading the begin record's payload */ if (drc->drc_rrd != NULL) { drc->drc_rrd->payload = buf; drc->drc_rrd->payload_size = len; drc->drc_rrd->bytes_read = drc->drc_bytes_read; } } else { ASSERT3P(buf, ==, NULL); } drc->drc_prev_cksum = drc->drc_cksum; drc->drc_next_rrd = kmem_zalloc(sizeof (*drc->drc_next_rrd), KM_SLEEP); err = receive_read(drc, sizeof (drc->drc_next_rrd->header), &drc->drc_next_rrd->header); drc->drc_next_rrd->bytes_read = drc->drc_bytes_read; if (err != 0) { kmem_free(drc->drc_next_rrd, sizeof (*drc->drc_next_rrd)); drc->drc_next_rrd = NULL; return (err); } if (drc->drc_next_rrd->header.drr_type == DRR_BEGIN) { kmem_free(drc->drc_next_rrd, sizeof (*drc->drc_next_rrd)); drc->drc_next_rrd = NULL; return (SET_ERROR(EINVAL)); } /* * Note: checksum is of everything up to but not including the * checksum itself. */ ASSERT3U(offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), ==, sizeof (dmu_replay_record_t) - sizeof (zio_cksum_t)); receive_cksum(drc, offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), &drc->drc_next_rrd->header); zio_cksum_t cksum_orig = drc->drc_next_rrd->header.drr_u.drr_checksum.drr_checksum; zio_cksum_t *cksump = &drc->drc_next_rrd->header.drr_u.drr_checksum.drr_checksum; if (drc->drc_byteswap) byteswap_record(&drc->drc_next_rrd->header); if ((!ZIO_CHECKSUM_IS_ZERO(cksump)) && !ZIO_CHECKSUM_EQUAL(drc->drc_cksum, *cksump)) { kmem_free(drc->drc_next_rrd, sizeof (*drc->drc_next_rrd)); drc->drc_next_rrd = NULL; return (SET_ERROR(ECKSUM)); } receive_cksum(drc, sizeof (cksum_orig), &cksum_orig); return (0); } /* * Issue the prefetch reads for any necessary indirect blocks. * * We use the object ignore list to tell us whether or not to issue prefetches * for a given object. We do this for both correctness (in case the blocksize * of an object has changed) and performance (if the object doesn't exist, don't * needlessly try to issue prefetches). We also trim the list as we go through * the stream to prevent it from growing to an unbounded size. * * The object numbers within will always be in sorted order, and any write * records we see will also be in sorted order, but they're not sorted with * respect to each other (i.e. we can get several object records before * receiving each object's write records). As a result, once we've reached a * given object number, we can safely remove any reference to lower object * numbers in the ignore list. In practice, we receive up to 32 object records * before receiving write records, so the list can have up to 32 nodes in it. */ static void receive_read_prefetch(dmu_recv_cookie_t *drc, uint64_t object, uint64_t offset, uint64_t length) { if (!objlist_exists(drc->drc_ignore_objlist, object)) { dmu_prefetch(drc->drc_os, object, 1, offset, length, ZIO_PRIORITY_SYNC_READ); } } /* * Read records off the stream, issuing any necessary prefetches. */ static int receive_read_record(dmu_recv_cookie_t *drc) { int err; switch (drc->drc_rrd->header.drr_type) { case DRR_OBJECT: { struct drr_object *drro = &drc->drc_rrd->header.drr_u.drr_object; uint32_t size = DRR_OBJECT_PAYLOAD_SIZE(drro); void *buf = NULL; dmu_object_info_t doi; if (size != 0) buf = kmem_zalloc(size, KM_SLEEP); err = receive_read_payload_and_next_header(drc, size, buf); if (err != 0) { kmem_free(buf, size); return (err); } err = dmu_object_info(drc->drc_os, drro->drr_object, &doi); /* * See receive_read_prefetch for an explanation why we're * storing this object in the ignore_obj_list. */ if (err == ENOENT || err == EEXIST || (err == 0 && doi.doi_data_block_size != drro->drr_blksz)) { objlist_insert(drc->drc_ignore_objlist, drro->drr_object); err = 0; } return (err); } case DRR_FREEOBJECTS: { err = receive_read_payload_and_next_header(drc, 0, NULL); return (err); } case DRR_WRITE: { struct drr_write *drrw = &drc->drc_rrd->header.drr_u.drr_write; int size = DRR_WRITE_PAYLOAD_SIZE(drrw); abd_t *abd = abd_alloc_linear(size, B_FALSE); err = receive_read_payload_and_next_header(drc, size, abd_to_buf(abd)); if (err != 0) { abd_free(abd); return (err); } drc->drc_rrd->abd = abd; receive_read_prefetch(drc, drrw->drr_object, drrw->drr_offset, drrw->drr_logical_size); return (err); } case DRR_WRITE_EMBEDDED: { struct drr_write_embedded *drrwe = &drc->drc_rrd->header.drr_u.drr_write_embedded; uint32_t size = P2ROUNDUP(drrwe->drr_psize, 8); void *buf = kmem_zalloc(size, KM_SLEEP); err = receive_read_payload_and_next_header(drc, size, buf); if (err != 0) { kmem_free(buf, size); return (err); } receive_read_prefetch(drc, drrwe->drr_object, drrwe->drr_offset, drrwe->drr_length); return (err); } case DRR_FREE: case DRR_REDACT: { /* * It might be beneficial to prefetch indirect blocks here, but * we don't really have the data to decide for sure. */ err = receive_read_payload_and_next_header(drc, 0, NULL); return (err); } case DRR_END: { struct drr_end *drre = &drc->drc_rrd->header.drr_u.drr_end; if (!ZIO_CHECKSUM_EQUAL(drc->drc_prev_cksum, drre->drr_checksum)) return (SET_ERROR(ECKSUM)); return (0); } case DRR_SPILL: { struct drr_spill *drrs = &drc->drc_rrd->header.drr_u.drr_spill; int size = DRR_SPILL_PAYLOAD_SIZE(drrs); abd_t *abd = abd_alloc_linear(size, B_FALSE); err = receive_read_payload_and_next_header(drc, size, abd_to_buf(abd)); if (err != 0) abd_free(abd); else drc->drc_rrd->abd = abd; return (err); } case DRR_OBJECT_RANGE: { err = receive_read_payload_and_next_header(drc, 0, NULL); return (err); } default: return (SET_ERROR(EINVAL)); } } static void dprintf_drr(struct receive_record_arg *rrd, int err) { #ifdef ZFS_DEBUG switch (rrd->header.drr_type) { case DRR_OBJECT: { struct drr_object *drro = &rrd->header.drr_u.drr_object; dprintf("drr_type = OBJECT obj = %llu type = %u " "bonustype = %u blksz = %u bonuslen = %u cksumtype = %u " "compress = %u dn_slots = %u err = %d\n", (u_longlong_t)drro->drr_object, drro->drr_type, drro->drr_bonustype, drro->drr_blksz, drro->drr_bonuslen, drro->drr_checksumtype, drro->drr_compress, drro->drr_dn_slots, err); break; } case DRR_FREEOBJECTS: { struct drr_freeobjects *drrfo = &rrd->header.drr_u.drr_freeobjects; dprintf("drr_type = FREEOBJECTS firstobj = %llu " "numobjs = %llu err = %d\n", (u_longlong_t)drrfo->drr_firstobj, (u_longlong_t)drrfo->drr_numobjs, err); break; } case DRR_WRITE: { struct drr_write *drrw = &rrd->header.drr_u.drr_write; dprintf("drr_type = WRITE obj = %llu type = %u offset = %llu " "lsize = %llu cksumtype = %u flags = %u " "compress = %u psize = %llu err = %d\n", (u_longlong_t)drrw->drr_object, drrw->drr_type, (u_longlong_t)drrw->drr_offset, (u_longlong_t)drrw->drr_logical_size, drrw->drr_checksumtype, drrw->drr_flags, drrw->drr_compressiontype, (u_longlong_t)drrw->drr_compressed_size, err); break; } case DRR_WRITE_BYREF: { struct drr_write_byref *drrwbr = &rrd->header.drr_u.drr_write_byref; dprintf("drr_type = WRITE_BYREF obj = %llu offset = %llu " "length = %llu toguid = %llx refguid = %llx " "refobject = %llu refoffset = %llu cksumtype = %u " "flags = %u err = %d\n", (u_longlong_t)drrwbr->drr_object, (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, drrwbr->drr_checksumtype, drrwbr->drr_flags, err); break; } case DRR_WRITE_EMBEDDED: { struct drr_write_embedded *drrwe = &rrd->header.drr_u.drr_write_embedded; dprintf("drr_type = WRITE_EMBEDDED obj = %llu offset = %llu " "length = %llu compress = %u etype = %u lsize = %u " "psize = %u err = %d\n", (u_longlong_t)drrwe->drr_object, (u_longlong_t)drrwe->drr_offset, (u_longlong_t)drrwe->drr_length, drrwe->drr_compression, drrwe->drr_etype, drrwe->drr_lsize, drrwe->drr_psize, err); break; } case DRR_FREE: { struct drr_free *drrf = &rrd->header.drr_u.drr_free; dprintf("drr_type = FREE obj = %llu offset = %llu " "length = %lld err = %d\n", (u_longlong_t)drrf->drr_object, (u_longlong_t)drrf->drr_offset, (longlong_t)drrf->drr_length, err); break; } case DRR_SPILL: { struct drr_spill *drrs = &rrd->header.drr_u.drr_spill; dprintf("drr_type = SPILL obj = %llu length = %llu " "err = %d\n", (u_longlong_t)drrs->drr_object, (u_longlong_t)drrs->drr_length, err); break; } case DRR_OBJECT_RANGE: { struct drr_object_range *drror = &rrd->header.drr_u.drr_object_range; dprintf("drr_type = OBJECT_RANGE firstobj = %llu " "numslots = %llu flags = %u err = %d\n", (u_longlong_t)drror->drr_firstobj, (u_longlong_t)drror->drr_numslots, drror->drr_flags, err); break; } default: return; } #endif } /* * Commit the records to the pool. */ static int receive_process_record(struct receive_writer_arg *rwa, struct receive_record_arg *rrd) { int err; /* Processing in order, therefore bytes_read should be increasing. */ ASSERT3U(rrd->bytes_read, >=, rwa->bytes_read); rwa->bytes_read = rrd->bytes_read; /* We can only heal write records; other ones get ignored */ if (rwa->heal && rrd->header.drr_type != DRR_WRITE) { if (rrd->abd != NULL) { abd_free(rrd->abd); rrd->abd = NULL; } else if (rrd->payload != NULL) { kmem_free(rrd->payload, rrd->payload_size); rrd->payload = NULL; } return (0); } if (!rwa->heal && rrd->header.drr_type != DRR_WRITE) { err = flush_write_batch(rwa); if (err != 0) { if (rrd->abd != NULL) { abd_free(rrd->abd); rrd->abd = NULL; rrd->payload = NULL; } else if (rrd->payload != NULL) { kmem_free(rrd->payload, rrd->payload_size); rrd->payload = NULL; } return (err); } } switch (rrd->header.drr_type) { case DRR_OBJECT: { struct drr_object *drro = &rrd->header.drr_u.drr_object; err = receive_object(rwa, drro, rrd->payload); kmem_free(rrd->payload, rrd->payload_size); rrd->payload = NULL; break; } case DRR_FREEOBJECTS: { struct drr_freeobjects *drrfo = &rrd->header.drr_u.drr_freeobjects; err = receive_freeobjects(rwa, drrfo); break; } case DRR_WRITE: { err = receive_process_write_record(rwa, rrd); if (rwa->heal) { /* * If healing - always free the abd after processing */ abd_free(rrd->abd); rrd->abd = NULL; } else if (err != EAGAIN) { /* * On success, a non-healing * receive_process_write_record() returns * EAGAIN to indicate that we do not want to free * the rrd or arc_buf. */ ASSERT(err != 0); abd_free(rrd->abd); rrd->abd = NULL; } break; } case DRR_WRITE_EMBEDDED: { struct drr_write_embedded *drrwe = &rrd->header.drr_u.drr_write_embedded; err = receive_write_embedded(rwa, drrwe, rrd->payload); kmem_free(rrd->payload, rrd->payload_size); rrd->payload = NULL; break; } case DRR_FREE: { struct drr_free *drrf = &rrd->header.drr_u.drr_free; err = receive_free(rwa, drrf); break; } case DRR_SPILL: { struct drr_spill *drrs = &rrd->header.drr_u.drr_spill; err = receive_spill(rwa, drrs, rrd->abd); if (err != 0) abd_free(rrd->abd); rrd->abd = NULL; rrd->payload = NULL; break; } case DRR_OBJECT_RANGE: { struct drr_object_range *drror = &rrd->header.drr_u.drr_object_range; err = receive_object_range(rwa, drror); break; } case DRR_REDACT: { struct drr_redact *drrr = &rrd->header.drr_u.drr_redact; err = receive_redact(rwa, drrr); break; } default: err = (SET_ERROR(EINVAL)); } if (err != 0) dprintf_drr(rrd, err); return (err); } /* * dmu_recv_stream's worker thread; pull records off the queue, and then call * receive_process_record When we're done, signal the main thread and exit. */ static __attribute__((noreturn)) void receive_writer_thread(void *arg) { struct receive_writer_arg *rwa = arg; struct receive_record_arg *rrd; fstrans_cookie_t cookie = spl_fstrans_mark(); for (rrd = bqueue_dequeue(&rwa->q); !rrd->eos_marker; rrd = bqueue_dequeue(&rwa->q)) { /* * If there's an error, the main thread will stop putting things * on the queue, but we need to clear everything in it before we * can exit. */ int err = 0; if (rwa->err == 0) { err = receive_process_record(rwa, rrd); } else if (rrd->abd != NULL) { abd_free(rrd->abd); rrd->abd = NULL; rrd->payload = NULL; } else if (rrd->payload != NULL) { kmem_free(rrd->payload, rrd->payload_size); rrd->payload = NULL; } /* * EAGAIN indicates that this record has been saved (on * raw->write_batch), and will be used again, so we don't * free it. * When healing data we always need to free the record. */ if (err != EAGAIN || rwa->heal) { if (rwa->err == 0) rwa->err = err; kmem_free(rrd, sizeof (*rrd)); } } kmem_free(rrd, sizeof (*rrd)); if (rwa->heal) { zio_wait(rwa->heal_pio); } else { int err = flush_write_batch(rwa); if (rwa->err == 0) rwa->err = err; } mutex_enter(&rwa->mutex); rwa->done = B_TRUE; cv_signal(&rwa->cv); mutex_exit(&rwa->mutex); spl_fstrans_unmark(cookie); thread_exit(); } static int resume_check(dmu_recv_cookie_t *drc, nvlist_t *begin_nvl) { uint64_t val; objset_t *mos = dmu_objset_pool(drc->drc_os)->dp_meta_objset; uint64_t dsobj = dmu_objset_id(drc->drc_os); uint64_t resume_obj, resume_off; if (nvlist_lookup_uint64(begin_nvl, "resume_object", &resume_obj) != 0 || nvlist_lookup_uint64(begin_nvl, "resume_offset", &resume_off) != 0) { return (SET_ERROR(EINVAL)); } VERIFY0(zap_lookup(mos, dsobj, DS_FIELD_RESUME_OBJECT, sizeof (val), 1, &val)); if (resume_obj != val) return (SET_ERROR(EINVAL)); VERIFY0(zap_lookup(mos, dsobj, DS_FIELD_RESUME_OFFSET, sizeof (val), 1, &val)); if (resume_off != val) return (SET_ERROR(EINVAL)); return (0); } /* * Read in the stream's records, one by one, and apply them to the pool. There * are two threads involved; the thread that calls this function will spin up a * worker thread, read the records off the stream one by one, and issue * prefetches for any necessary indirect blocks. It will then push the records * onto an internal blocking queue. The worker thread will pull the records off * the queue, and actually write the data into the DMU. This way, the worker * thread doesn't have to wait for reads to complete, since everything it needs * (the indirect blocks) will be prefetched. * * NB: callers *must* call dmu_recv_end() if this succeeds. */ int dmu_recv_stream(dmu_recv_cookie_t *drc, offset_t *voffp) { int err = 0; struct receive_writer_arg *rwa = kmem_zalloc(sizeof (*rwa), KM_SLEEP); if (dsl_dataset_has_resume_receive_state(drc->drc_ds)) { uint64_t bytes = 0; (void) zap_lookup(drc->drc_ds->ds_dir->dd_pool->dp_meta_objset, drc->drc_ds->ds_object, DS_FIELD_RESUME_BYTES, sizeof (bytes), 1, &bytes); drc->drc_bytes_read += bytes; } drc->drc_ignore_objlist = objlist_create(); /* 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); ASSERT(dsl_dataset_phys(drc->drc_ds)->ds_flags & DS_FLAG_INCONSISTENT); ASSERT0(drc->drc_os->os_encrypted && (drc->drc_featureflags & DMU_BACKUP_FEATURE_EMBED_DATA)); /* handle DSL encryption key payload */ if (drc->drc_featureflags & DMU_BACKUP_FEATURE_RAW) { nvlist_t *keynvl = NULL; ASSERT(drc->drc_os->os_encrypted); ASSERT(drc->drc_raw); err = nvlist_lookup_nvlist(drc->drc_begin_nvl, "crypt_keydata", &keynvl); if (err != 0) goto out; if (!drc->drc_heal) { /* * If this is a new dataset we set the key immediately. * Otherwise we don't want to change the key until we * are sure the rest of the receive succeeded so we * stash the keynvl away until then. */ err = dsl_crypto_recv_raw(spa_name(drc->drc_os->os_spa), drc->drc_ds->ds_object, drc->drc_fromsnapobj, drc->drc_drrb->drr_type, keynvl, drc->drc_newfs); if (err != 0) goto out; } /* see comment in dmu_recv_end_sync() */ drc->drc_ivset_guid = 0; (void) nvlist_lookup_uint64(keynvl, "to_ivset_guid", &drc->drc_ivset_guid); if (!drc->drc_newfs) drc->drc_keynvl = fnvlist_dup(keynvl); } if (drc->drc_featureflags & DMU_BACKUP_FEATURE_RESUMING) { err = resume_check(drc, drc->drc_begin_nvl); if (err != 0) goto out; } /* * For compatibility with recursive send streams, we do this here, * rather than in dmu_recv_begin. If we pull the next header too * early, and it's the END record, we break the `recv_skip` logic. */ if (drc->drc_drr_begin->drr_payloadlen == 0) { err = receive_read_payload_and_next_header(drc, 0, NULL); if (err != 0) goto out; } /* * If we failed before this point we will clean up any new resume * state that was created. Now that we've gotten past the initial * checks we are ok to retain that resume state. */ drc->drc_should_save = B_TRUE; (void) bqueue_init(&rwa->q, zfs_recv_queue_ff, MAX(zfs_recv_queue_length, 2 * zfs_max_recordsize), offsetof(struct receive_record_arg, node)); cv_init(&rwa->cv, NULL, CV_DEFAULT, NULL); mutex_init(&rwa->mutex, NULL, MUTEX_DEFAULT, NULL); rwa->os = drc->drc_os; rwa->byteswap = drc->drc_byteswap; rwa->heal = drc->drc_heal; rwa->tofs = drc->drc_tofs; rwa->resumable = drc->drc_resumable; rwa->raw = drc->drc_raw; rwa->spill = drc->drc_spill; rwa->full = (drc->drc_drr_begin->drr_u.drr_begin.drr_fromguid == 0); rwa->os->os_raw_receive = drc->drc_raw; if (drc->drc_heal) { rwa->heal_pio = zio_root(drc->drc_os->os_spa, NULL, NULL, ZIO_FLAG_GODFATHER); } list_create(&rwa->write_batch, sizeof (struct receive_record_arg), offsetof(struct receive_record_arg, node.bqn_node)); (void) thread_create(NULL, 0, receive_writer_thread, rwa, 0, curproc, TS_RUN, minclsyspri); /* * We're reading rwa->err without locks, which is safe since we are the * only reader, and the worker thread is the only writer. It's ok if we * miss a write for an iteration or two of the loop, since the writer * thread will keep freeing records we send it until we send it an eos * marker. * * We can leave this loop in 3 ways: First, if rwa->err is * non-zero. In that case, the writer thread will free the rrd we just * pushed. Second, if we're interrupted; in that case, either it's the * first loop and drc->drc_rrd was never allocated, or it's later, and * drc->drc_rrd has been handed off to the writer thread who will free * it. Finally, if receive_read_record fails or we're at the end of the * stream, then we free drc->drc_rrd and exit. */ while (rwa->err == 0) { if (issig(JUSTLOOKING) && issig(FORREAL)) { err = SET_ERROR(EINTR); break; } ASSERT3P(drc->drc_rrd, ==, NULL); drc->drc_rrd = drc->drc_next_rrd; drc->drc_next_rrd = NULL; /* Allocates and loads header into drc->drc_next_rrd */ err = receive_read_record(drc); if (drc->drc_rrd->header.drr_type == DRR_END || err != 0) { kmem_free(drc->drc_rrd, sizeof (*drc->drc_rrd)); drc->drc_rrd = NULL; break; } bqueue_enqueue(&rwa->q, drc->drc_rrd, sizeof (struct receive_record_arg) + drc->drc_rrd->payload_size); drc->drc_rrd = NULL; } ASSERT3P(drc->drc_rrd, ==, NULL); drc->drc_rrd = kmem_zalloc(sizeof (*drc->drc_rrd), KM_SLEEP); drc->drc_rrd->eos_marker = B_TRUE; bqueue_enqueue_flush(&rwa->q, drc->drc_rrd, 1); mutex_enter(&rwa->mutex); while (!rwa->done) { /* * We need to use cv_wait_sig() so that any process that may * be sleeping here can still fork. */ (void) cv_wait_sig(&rwa->cv, &rwa->mutex); } mutex_exit(&rwa->mutex); /* * If we are receiving a full stream as a clone, all object IDs which * are greater than the maximum ID referenced in the stream are * by definition unused and must be freed. */ if (drc->drc_clone && drc->drc_drrb->drr_fromguid == 0) { uint64_t obj = rwa->max_object + 1; int free_err = 0; int next_err = 0; while (next_err == 0) { free_err = dmu_free_long_object(rwa->os, obj); if (free_err != 0 && free_err != ENOENT) break; next_err = dmu_object_next(rwa->os, &obj, FALSE, 0); } if (err == 0) { if (free_err != 0 && free_err != ENOENT) err = free_err; else if (next_err != ESRCH) err = next_err; } } cv_destroy(&rwa->cv); mutex_destroy(&rwa->mutex); bqueue_destroy(&rwa->q); list_destroy(&rwa->write_batch); if (err == 0) err = rwa->err; out: /* * If we hit an error before we started the receive_writer_thread * we need to clean up the next_rrd we create by processing the * DRR_BEGIN record. */ if (drc->drc_next_rrd != NULL) kmem_free(drc->drc_next_rrd, sizeof (*drc->drc_next_rrd)); /* * The objset will be invalidated by dmu_recv_end() when we do * dsl_dataset_clone_swap_sync_impl(). */ drc->drc_os = NULL; kmem_free(rwa, sizeof (*rwa)); nvlist_free(drc->drc_begin_nvl); if (err != 0) { /* * Clean up references. If receive is not resumable, * destroy what we created, so we don't leave it in * the inconsistent state. */ dmu_recv_cleanup_ds(drc); nvlist_free(drc->drc_keynvl); } objlist_destroy(drc->drc_ignore_objlist); drc->drc_ignore_objlist = NULL; *voffp = drc->drc_voff; 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_heal) { error = 0; } else 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); } } if (drc->drc_keynvl != NULL) { error = dsl_crypto_recv_raw_key_check(drc->drc_ds, drc->drc_keynvl, tx); 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, drc->drc_proc); 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, drc->drc_proc); } 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); boolean_t encrypted = drc->drc_ds->ds_dir->dd_crypto_obj != 0; uint64_t newsnapobj = 0; spa_history_log_internal_ds(drc->drc_ds, "finish receiving", tx, "snap=%s", drc->drc_tosnap); drc->drc_ds->ds_objset->os_raw_receive = B_FALSE; if (drc->drc_heal) { if (drc->drc_keynvl != NULL) { nvlist_free(drc->drc_keynvl); drc->drc_keynvl = NULL; } } else 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); } } if (drc->drc_keynvl != NULL) { dsl_crypto_recv_raw_key_sync(drc->drc_ds, drc->drc_keynvl, tx); nvlist_free(drc->drc_keynvl); drc->drc_keynvl = NULL; } VERIFY3P(drc->drc_ds->ds_prev, ==, origin_head->ds_prev); dsl_dataset_clone_swap_sync_impl(drc->drc_ds, origin_head, tx); /* * The objset was evicted by dsl_dataset_clone_swap_sync_impl, * so drc_os is no longer valid. */ drc->drc_os = NULL; 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; newsnapobj = dsl_dataset_phys(origin_head)->ds_prev_snap_obj; dsl_dataset_rele(origin_head, FTAG); dsl_destroy_head_sync_impl(drc->drc_ds, tx); if (drc->drc_owner != NULL) VERIFY3P(origin_head->ds_owner, ==, drc->drc_owner); } else { dsl_dataset_t *ds = drc->drc_ds; dsl_dataset_snapshot_sync_impl(ds, drc->drc_tosnap, tx); /* set snapshot's creation time and guid */ dmu_buf_will_dirty(ds->ds_prev->ds_dbuf, tx); dsl_dataset_phys(ds->ds_prev)->ds_creation_time = drc->drc_drrb->drr_creation_time; dsl_dataset_phys(ds->ds_prev)->ds_guid = drc->drc_drrb->drr_toguid; dsl_dataset_phys(ds->ds_prev)->ds_flags &= ~DS_FLAG_INCONSISTENT; dmu_buf_will_dirty(ds->ds_dbuf, tx); dsl_dataset_phys(ds)->ds_flags &= ~DS_FLAG_INCONSISTENT; if (dsl_dataset_has_resume_receive_state(ds)) { (void) zap_remove(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_FROMGUID, tx); (void) zap_remove(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_OBJECT, tx); (void) zap_remove(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_OFFSET, tx); (void) zap_remove(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_BYTES, tx); (void) zap_remove(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_TOGUID, tx); (void) zap_remove(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_TONAME, tx); (void) zap_remove(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_REDACT_BOOKMARK_SNAPS, tx); } newsnapobj = dsl_dataset_phys(drc->drc_ds)->ds_prev_snap_obj; } /* * If this is a raw receive, the crypt_keydata nvlist will include * a to_ivset_guid for us to set on the new snapshot. This value * will override the value generated by the snapshot code. However, * this value may not be present, because older implementations of * the raw send code did not include this value, and we are still * allowed to receive them if the zfs_disable_ivset_guid_check * tunable is set, in which case we will leave the newly-generated * value. */ if (!drc->drc_heal && drc->drc_raw && drc->drc_ivset_guid != 0) { dmu_object_zapify(dp->dp_meta_objset, newsnapobj, DMU_OT_DSL_DATASET, tx); VERIFY0(zap_update(dp->dp_meta_objset, newsnapobj, DS_FIELD_IVSET_GUID, sizeof (uint64_t), 1, &drc->drc_ivset_guid, tx)); } /* * 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. Since the dataset may be destroyed * at this point (and therefore won't have a valid pointer to the spa) * we release the key mapping manually here while we do have a valid * pointer, if it exists. */ if (!drc->drc_raw && encrypted) { (void) spa_keystore_remove_mapping(dmu_tx_pool(tx)->dp_spa, drc->drc_ds->ds_object, drc->drc_ds); } dsl_dataset_disown(drc->drc_ds, 0, dmu_recv_tag); drc->drc_ds = NULL; } static int dmu_recv_end_modified_blocks = 3; static int dmu_recv_existing_end(dmu_recv_cookie_t *drc) { #ifdef _KERNEL /* * We will be destroying the ds; make sure its origin is unmounted if * necessary. */ char name[ZFS_MAX_DATASET_NAME_LEN]; dsl_dataset_name(drc->drc_ds, name); zfs_destroy_unmount_origin(name); #endif return (dsl_sync_task(drc->drc_tofs, dmu_recv_end_check, dmu_recv_end_sync, drc, dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL)); } static int dmu_recv_new_end(dmu_recv_cookie_t *drc) { return (dsl_sync_task(drc->drc_tofs, dmu_recv_end_check, dmu_recv_end_sync, drc, dmu_recv_end_modified_blocks, ZFS_SPACE_CHECK_NORMAL)); } int dmu_recv_end(dmu_recv_cookie_t *drc, void *owner) { int error; drc->drc_owner = owner; if (drc->drc_newfs) error = dmu_recv_new_end(drc); else error = dmu_recv_existing_end(drc); if (error != 0) { dmu_recv_cleanup_ds(drc); nvlist_free(drc->drc_keynvl); } else if (!drc->drc_heal) { if (drc->drc_newfs) { zvol_create_minor(drc->drc_tofs); } char *snapname = kmem_asprintf("%s@%s", drc->drc_tofs, drc->drc_tosnap); zvol_create_minor(snapname); kmem_strfree(snapname); } return (error); } /* * Return TRUE if this objset is currently being received into. */ boolean_t dmu_objset_is_receiving(objset_t *os) { return (os->os_dsl_dataset != NULL && os->os_dsl_dataset->ds_owner == dmu_recv_tag); } ZFS_MODULE_PARAM(zfs_recv, zfs_recv_, queue_length, UINT, ZMOD_RW, "Maximum receive queue length"); ZFS_MODULE_PARAM(zfs_recv, zfs_recv_, queue_ff, UINT, ZMOD_RW, "Receive queue fill fraction"); ZFS_MODULE_PARAM(zfs_recv, zfs_recv_, write_batch_size, UINT, ZMOD_RW, "Maximum amount of writes to batch into one transaction"); ZFS_MODULE_PARAM(zfs_recv, zfs_recv_, best_effort_corrective, INT, ZMOD_RW, "Ignore errors during corrective receive"); /* END CSTYLED */ diff --git a/module/zfs/spa_errlog.c b/module/zfs/spa_errlog.c index 44950a769d3b..31719063a227 100644 --- a/module/zfs/spa_errlog.c +++ b/module/zfs/spa_errlog.c @@ -1,1360 +1,1465 @@ /* * 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 https://opensource.org/licenses/CDDL-1.0. * 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) 2006, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright (c) 2013, 2014, Delphix. All rights reserved. * Copyright (c) 2019 Datto Inc. * Copyright (c) 2021, 2022, George Amanakis. All rights reserved. */ /* * Routines to manage the on-disk persistent error log. * * Each pool stores a log of all logical data errors seen during normal * operation. This is actually the union of two distinct logs: the last log, * and the current log. All errors seen are logged to the current log. When a * scrub completes, the current log becomes the last log, the last log is thrown * out, and the current log is reinitialized. This way, if an error is somehow * corrected, a new scrub will show that it no longer exists, and will be * deleted from the log when the scrub completes. * * The log is stored using a ZAP object whose key is a string form of the * zbookmark_phys tuple (objset, object, level, blkid), and whose contents is an * optional 'objset:object' human-readable string describing the data. When an * error is first logged, this string will be empty, indicating that no name is * known. This prevents us from having to issue a potentially large amount of * I/O to discover the object name during an error path. Instead, we do the * calculation when the data is requested, storing the result so future queries * will be faster. * * If the head_errlog feature is enabled, a different on-disk format is used. * The error log of each head dataset is stored separately in the zap object * and keyed by the head id. This enables listing every dataset affected in * userland. In order to be able to track whether an error block has been * modified or added to snapshots since it was marked as an error, a new tuple * is introduced: zbookmark_err_phys_t. It allows the storage of the birth * transaction group of an error block on-disk. The birth transaction group is * used by check_filesystem() to assess whether this block was freed, * re-written or added to a snapshot since its marking as an error. * * This log is then shipped into an nvlist where the key is the dataset name and * the value is the object name. Userland is then responsible for uniquifying * this list and displaying it to the user. */ #include #include #include #include #include #include #include #include #include #define NAME_MAX_LEN 64 typedef struct clones { uint64_t clone_ds; list_node_t node; } clones_t; /* * spa_upgrade_errlog_limit : A zfs module parameter that controls the number * of on-disk error log entries that will be converted to the new * format when enabling head_errlog. Defaults to 0 which converts * all log entries. */ static uint_t spa_upgrade_errlog_limit = 0; /* * Convert a bookmark to a string. */ static void bookmark_to_name(zbookmark_phys_t *zb, char *buf, size_t len) { (void) snprintf(buf, len, "%llx:%llx:%llx:%llx", (u_longlong_t)zb->zb_objset, (u_longlong_t)zb->zb_object, (u_longlong_t)zb->zb_level, (u_longlong_t)zb->zb_blkid); } /* * Convert an err_phys to a string. */ static void errphys_to_name(zbookmark_err_phys_t *zep, char *buf, size_t len) { (void) snprintf(buf, len, "%llx:%llx:%llx:%llx", (u_longlong_t)zep->zb_object, (u_longlong_t)zep->zb_level, (u_longlong_t)zep->zb_blkid, (u_longlong_t)zep->zb_birth); } /* * Convert a string to a err_phys. */ static void name_to_errphys(char *buf, zbookmark_err_phys_t *zep) { zep->zb_object = zfs_strtonum(buf, &buf); ASSERT(*buf == ':'); zep->zb_level = (int)zfs_strtonum(buf + 1, &buf); ASSERT(*buf == ':'); zep->zb_blkid = zfs_strtonum(buf + 1, &buf); ASSERT(*buf == ':'); zep->zb_birth = zfs_strtonum(buf + 1, &buf); ASSERT(*buf == '\0'); } /* * Convert a string to a bookmark. */ static void name_to_bookmark(char *buf, zbookmark_phys_t *zb) { zb->zb_objset = zfs_strtonum(buf, &buf); ASSERT(*buf == ':'); zb->zb_object = zfs_strtonum(buf + 1, &buf); ASSERT(*buf == ':'); zb->zb_level = (int)zfs_strtonum(buf + 1, &buf); ASSERT(*buf == ':'); zb->zb_blkid = zfs_strtonum(buf + 1, &buf); ASSERT(*buf == '\0'); } #ifdef _KERNEL static void zep_to_zb(uint64_t dataset, zbookmark_err_phys_t *zep, zbookmark_phys_t *zb) { zb->zb_objset = dataset; zb->zb_object = zep->zb_object; zb->zb_level = zep->zb_level; zb->zb_blkid = zep->zb_blkid; } #endif static void name_to_object(char *buf, uint64_t *obj) { *obj = zfs_strtonum(buf, &buf); ASSERT(*buf == '\0'); } /* * Retrieve the head filesystem. */ static int get_head_ds(spa_t *spa, uint64_t dsobj, uint64_t *head_ds) { dsl_dataset_t *ds; int error = dsl_dataset_hold_obj_flags(spa->spa_dsl_pool, dsobj, DS_HOLD_FLAG_DECRYPT, FTAG, &ds); if (error != 0) return (error); ASSERT(head_ds); *head_ds = dsl_dir_phys(ds->ds_dir)->dd_head_dataset_obj; dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); return (error); } /* * Log an uncorrectable error to the persistent error log. We add it to the * spa's list of pending errors. The changes are actually synced out to disk * during spa_errlog_sync(). */ void spa_log_error(spa_t *spa, const zbookmark_phys_t *zb, const uint64_t *birth) { spa_error_entry_t search; spa_error_entry_t *new; avl_tree_t *tree; avl_index_t where; /* * If we are trying to import a pool, ignore any errors, as we won't be * writing to the pool any time soon. */ if (spa_load_state(spa) == SPA_LOAD_TRYIMPORT) return; mutex_enter(&spa->spa_errlist_lock); /* * If we have had a request to rotate the log, log it to the next list * instead of the current one. */ if (spa->spa_scrub_active || spa->spa_scrub_finished) tree = &spa->spa_errlist_scrub; else tree = &spa->spa_errlist_last; search.se_bookmark = *zb; if (avl_find(tree, &search, &where) != NULL) { mutex_exit(&spa->spa_errlist_lock); return; } new = kmem_zalloc(sizeof (spa_error_entry_t), KM_SLEEP); new->se_bookmark = *zb; /* * If the head_errlog feature is enabled, store the birth txg now. In * case the file is deleted before spa_errlog_sync() runs, we will not * be able to retrieve the birth txg. */ if (spa_feature_is_enabled(spa, SPA_FEATURE_HEAD_ERRLOG)) { new->se_zep.zb_object = zb->zb_object; new->se_zep.zb_level = zb->zb_level; new->se_zep.zb_blkid = zb->zb_blkid; /* * birth may end up being NULL, e.g. in zio_done(). We * will handle this in process_error_block(). */ if (birth != NULL) new->se_zep.zb_birth = *birth; } avl_insert(tree, new, where); mutex_exit(&spa->spa_errlist_lock); } #ifdef _KERNEL static int find_birth_txg(dsl_dataset_t *ds, zbookmark_err_phys_t *zep, uint64_t *birth_txg) { objset_t *os; int error = dmu_objset_from_ds(ds, &os); if (error != 0) return (error); dnode_t *dn; blkptr_t bp; error = dnode_hold(os, zep->zb_object, FTAG, &dn); if (error != 0) return (error); rw_enter(&dn->dn_struct_rwlock, RW_READER); error = dbuf_dnode_findbp(dn, zep->zb_level, zep->zb_blkid, &bp, NULL, NULL); if (error == 0 && BP_IS_HOLE(&bp)) error = SET_ERROR(ENOENT); *birth_txg = bp.blk_birth; rw_exit(&dn->dn_struct_rwlock); dnode_rele(dn, FTAG); return (error); } /* * Copy the bookmark to the end of the user-space buffer which starts at * uaddr and has *count unused entries, and decrement *count by 1. */ static int copyout_entry(const zbookmark_phys_t *zb, void *uaddr, uint64_t *count) { if (*count == 0) return (SET_ERROR(ENOMEM)); *count -= 1; if (copyout(zb, (char *)uaddr + (*count) * sizeof (zbookmark_phys_t), sizeof (zbookmark_phys_t)) != 0) return (SET_ERROR(EFAULT)); return (0); } /* * Each time the error block is referenced by a snapshot or clone, add a * zbookmark_phys_t entry to the userspace array at uaddr. The array is * filled from the back and the in-out parameter *count is modified to be the * number of unused entries at the beginning of the array. */ static int check_filesystem(spa_t *spa, uint64_t head_ds, zbookmark_err_phys_t *zep, void *uaddr, uint64_t *count, list_t *clones_list) { dsl_dataset_t *ds; dsl_pool_t *dp = spa->spa_dsl_pool; int error = dsl_dataset_hold_obj_flags(dp, head_ds, DS_HOLD_FLAG_DECRYPT, FTAG, &ds); if (error != 0) return (error); uint64_t latest_txg; uint64_t txg_to_consider = spa->spa_syncing_txg; boolean_t check_snapshot = B_TRUE; error = find_birth_txg(ds, zep, &latest_txg); /* * If find_birth_txg() errors out otherwise, let txg_to_consider be * equal to the spa's syncing txg: if check_filesystem() errors out * then affected snapshots or clones will not be checked. */ if (error == 0 && zep->zb_birth == latest_txg) { /* Block neither free nor rewritten. */ zbookmark_phys_t zb; zep_to_zb(head_ds, zep, &zb); error = copyout_entry(&zb, uaddr, count); if (error != 0) { dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); return (error); } check_snapshot = B_FALSE; } else if (error == 0) { txg_to_consider = latest_txg; } /* * Retrieve the number of snapshots if the dataset is not a snapshot. */ uint64_t snap_count = 0; if (dsl_dataset_phys(ds)->ds_snapnames_zapobj != 0) { error = zap_count(spa->spa_meta_objset, dsl_dataset_phys(ds)->ds_snapnames_zapobj, &snap_count); if (error != 0) { dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); return (error); } } if (snap_count == 0) { /* Filesystem without snapshots. */ dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); return (0); } uint64_t *snap_obj_array = kmem_zalloc(snap_count * sizeof (uint64_t), KM_SLEEP); int aff_snap_count = 0; uint64_t snap_obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; uint64_t snap_obj_txg = dsl_dataset_phys(ds)->ds_prev_snap_txg; uint64_t zap_clone = dsl_dir_phys(ds->ds_dir)->dd_clones; dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); /* Check only snapshots created from this file system. */ while (snap_obj != 0 && zep->zb_birth < snap_obj_txg && snap_obj_txg <= txg_to_consider) { error = dsl_dataset_hold_obj_flags(dp, snap_obj, DS_HOLD_FLAG_DECRYPT, FTAG, &ds); if (error != 0) goto out; if (dsl_dir_phys(ds->ds_dir)->dd_head_dataset_obj != head_ds) { snap_obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; snap_obj_txg = dsl_dataset_phys(ds)->ds_prev_snap_txg; dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); continue; } boolean_t affected = B_TRUE; if (check_snapshot) { uint64_t blk_txg; error = find_birth_txg(ds, zep, &blk_txg); affected = (error == 0 && zep->zb_birth == blk_txg); } /* Report errors in snapshots. */ if (affected) { snap_obj_array[aff_snap_count] = snap_obj; aff_snap_count++; zbookmark_phys_t zb; zep_to_zb(snap_obj, zep, &zb); error = copyout_entry(&zb, uaddr, count); if (error != 0) { dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); goto out; } } snap_obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; snap_obj_txg = dsl_dataset_phys(ds)->ds_prev_snap_txg; dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); } if (zap_clone == 0 || aff_snap_count == 0) return (0); /* Check clones. */ zap_cursor_t *zc; zap_attribute_t *za; zc = kmem_zalloc(sizeof (zap_cursor_t), KM_SLEEP); za = kmem_zalloc(sizeof (zap_attribute_t), KM_SLEEP); for (zap_cursor_init(zc, spa->spa_meta_objset, zap_clone); zap_cursor_retrieve(zc, za) == 0; zap_cursor_advance(zc)) { dsl_dataset_t *clone; error = dsl_dataset_hold_obj_flags(dp, za->za_first_integer, DS_HOLD_FLAG_DECRYPT, FTAG, &clone); if (error != 0) break; /* * Only clones whose origins were affected could also * have affected snapshots. */ boolean_t found = B_FALSE; for (int i = 0; i < snap_count; i++) { if (dsl_dir_phys(clone->ds_dir)->dd_origin_obj == snap_obj_array[i]) found = B_TRUE; } dsl_dataset_rele_flags(clone, DS_HOLD_FLAG_DECRYPT, FTAG); if (!found) continue; clones_t *ct = kmem_zalloc(sizeof (*ct), KM_SLEEP); ct->clone_ds = za->za_first_integer; list_insert_tail(clones_list, ct); } zap_cursor_fini(zc); kmem_free(za, sizeof (*za)); kmem_free(zc, sizeof (*zc)); out: kmem_free(snap_obj_array, sizeof (*snap_obj_array)); return (error); } static int find_top_affected_fs(spa_t *spa, uint64_t head_ds, zbookmark_err_phys_t *zep, uint64_t *top_affected_fs) { uint64_t oldest_dsobj; int error = dsl_dataset_oldest_snapshot(spa, head_ds, zep->zb_birth, &oldest_dsobj); if (error != 0) return (error); dsl_dataset_t *ds; error = dsl_dataset_hold_obj_flags(spa->spa_dsl_pool, oldest_dsobj, DS_HOLD_FLAG_DECRYPT, FTAG, &ds); if (error != 0) return (error); *top_affected_fs = dsl_dir_phys(ds->ds_dir)->dd_head_dataset_obj; dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); return (0); } static int process_error_block(spa_t *spa, uint64_t head_ds, zbookmark_err_phys_t *zep, void *uaddr, uint64_t *count) { /* * If zb_birth == 0 or head_ds == 0 it means we failed to retrieve the * birth txg or the head filesystem of the block pointer. This may * happen e.g. when an encrypted filesystem is not mounted or when * the key is not loaded. In this case do not proceed to * check_filesystem(), instead do the accounting here. */ if (zep->zb_birth == 0 || head_ds == 0) { zbookmark_phys_t zb; zep_to_zb(head_ds, zep, &zb); int error = copyout_entry(&zb, uaddr, count); if (error != 0) { return (error); } return (0); } uint64_t top_affected_fs; + uint64_t init_count = *count; int error = find_top_affected_fs(spa, head_ds, zep, &top_affected_fs); if (error == 0) { clones_t *ct; list_t clones_list; list_create(&clones_list, sizeof (clones_t), offsetof(clones_t, node)); error = check_filesystem(spa, top_affected_fs, zep, uaddr, count, &clones_list); while ((ct = list_remove_head(&clones_list)) != NULL) { error = check_filesystem(spa, ct->clone_ds, zep, uaddr, count, &clones_list); kmem_free(ct, sizeof (*ct)); if (error) { while (!list_is_empty(&clones_list)) { ct = list_remove_head(&clones_list); kmem_free(ct, sizeof (*ct)); } break; } } list_destroy(&clones_list); } + if (error == 0 && init_count == *count) { + /* + * If we reach this point, no errors have been detected + * in the checked filesystems/snapshots. Before returning mark + * the error block to be removed from the error lists and logs. + */ + zbookmark_phys_t zb; + zep_to_zb(head_ds, zep, &zb); + spa_remove_error(spa, &zb, &zep->zb_birth); + } return (error); } #endif /* * If a healed bookmark matches an entry in the error log we stash it in a tree * so that we can later remove the related log entries in sync context. */ static void -spa_add_healed_error(spa_t *spa, uint64_t obj, zbookmark_phys_t *healed_zb) +spa_add_healed_error(spa_t *spa, uint64_t obj, zbookmark_phys_t *healed_zb, + const uint64_t *birth) { char name[NAME_MAX_LEN]; if (obj == 0) return; - bookmark_to_name(healed_zb, name, sizeof (name)); - mutex_enter(&spa->spa_errlog_lock); - if (zap_contains(spa->spa_meta_objset, obj, name) == 0) { - /* - * Found an error matching healed zb, add zb to our - * tree of healed errors - */ - avl_tree_t *tree = &spa->spa_errlist_healed; - spa_error_entry_t search; - spa_error_entry_t *new; - avl_index_t where; - search.se_bookmark = *healed_zb; - mutex_enter(&spa->spa_errlist_lock); - if (avl_find(tree, &search, &where) != NULL) { - mutex_exit(&spa->spa_errlist_lock); - mutex_exit(&spa->spa_errlog_lock); - return; + boolean_t held_list = B_FALSE; + boolean_t held_log = B_FALSE; + + if (!spa_feature_is_enabled(spa, SPA_FEATURE_HEAD_ERRLOG)) { + bookmark_to_name(healed_zb, name, sizeof (name)); + + if (zap_contains(spa->spa_meta_objset, healed_zb->zb_objset, + name) == 0) { + if (!MUTEX_HELD(&spa->spa_errlog_lock)) { + mutex_enter(&spa->spa_errlog_lock); + held_log = B_TRUE; + } + + /* + * Found an error matching healed zb, add zb to our + * tree of healed errors + */ + avl_tree_t *tree = &spa->spa_errlist_healed; + spa_error_entry_t search; + spa_error_entry_t *new; + avl_index_t where; + search.se_bookmark = *healed_zb; + if (!MUTEX_HELD(&spa->spa_errlist_lock)) { + mutex_enter(&spa->spa_errlist_lock); + held_list = B_TRUE; + } + if (avl_find(tree, &search, &where) != NULL) { + if (held_list) + mutex_exit(&spa->spa_errlist_lock); + if (held_log) + mutex_exit(&spa->spa_errlog_lock); + return; + } + new = kmem_zalloc(sizeof (spa_error_entry_t), KM_SLEEP); + new->se_bookmark = *healed_zb; + avl_insert(tree, new, where); + if (held_list) + mutex_exit(&spa->spa_errlist_lock); + if (held_log) + mutex_exit(&spa->spa_errlog_lock); } - new = kmem_zalloc(sizeof (spa_error_entry_t), KM_SLEEP); - new->se_bookmark = *healed_zb; - avl_insert(tree, new, where); - mutex_exit(&spa->spa_errlist_lock); + return; } - mutex_exit(&spa->spa_errlog_lock); + + zbookmark_err_phys_t healed_zep; + healed_zep.zb_object = healed_zb->zb_object; + healed_zep.zb_level = healed_zb->zb_level; + healed_zep.zb_blkid = healed_zb->zb_blkid; + + if (birth != NULL) + healed_zep.zb_birth = *birth; + else + healed_zep.zb_birth = 0; + + errphys_to_name(&healed_zep, name, sizeof (name)); + + zap_cursor_t zc; + zap_attribute_t za; + for (zap_cursor_init(&zc, spa->spa_meta_objset, spa->spa_errlog_last); + zap_cursor_retrieve(&zc, &za) == 0; zap_cursor_advance(&zc)) { + if (zap_contains(spa->spa_meta_objset, za.za_first_integer, + name) == 0) { + if (!MUTEX_HELD(&spa->spa_errlog_lock)) { + mutex_enter(&spa->spa_errlog_lock); + held_log = B_TRUE; + } + + avl_tree_t *tree = &spa->spa_errlist_healed; + spa_error_entry_t search; + spa_error_entry_t *new; + avl_index_t where; + search.se_bookmark = *healed_zb; + + if (!MUTEX_HELD(&spa->spa_errlist_lock)) { + mutex_enter(&spa->spa_errlist_lock); + held_list = B_TRUE; + } + + if (avl_find(tree, &search, &where) != NULL) { + if (held_list) + mutex_exit(&spa->spa_errlist_lock); + if (held_log) + mutex_exit(&spa->spa_errlog_lock); + continue; + } + new = kmem_zalloc(sizeof (spa_error_entry_t), KM_SLEEP); + new->se_bookmark = *healed_zb; + new->se_zep = healed_zep; + avl_insert(tree, new, where); + + if (held_list) + mutex_exit(&spa->spa_errlist_lock); + if (held_log) + mutex_exit(&spa->spa_errlog_lock); + } + } + zap_cursor_fini(&zc); } /* * If this error exists in the given tree remove it. */ static void remove_error_from_list(spa_t *spa, avl_tree_t *t, const zbookmark_phys_t *zb) { spa_error_entry_t search, *found; avl_index_t where; mutex_enter(&spa->spa_errlist_lock); search.se_bookmark = *zb; if ((found = avl_find(t, &search, &where)) != NULL) { avl_remove(t, found); kmem_free(found, sizeof (spa_error_entry_t)); } mutex_exit(&spa->spa_errlist_lock); } /* * Removes all of the recv healed errors from both on-disk error logs */ static void spa_remove_healed_errors(spa_t *spa, avl_tree_t *s, avl_tree_t *l, dmu_tx_t *tx) { char name[NAME_MAX_LEN]; spa_error_entry_t *se; void *cookie = NULL; ASSERT(MUTEX_HELD(&spa->spa_errlog_lock)); while ((se = avl_destroy_nodes(&spa->spa_errlist_healed, &cookie)) != NULL) { remove_error_from_list(spa, s, &se->se_bookmark); remove_error_from_list(spa, l, &se->se_bookmark); - bookmark_to_name(&se->se_bookmark, name, sizeof (name)); kmem_free(se, sizeof (spa_error_entry_t)); - (void) zap_remove(spa->spa_meta_objset, - spa->spa_errlog_last, name, tx); - (void) zap_remove(spa->spa_meta_objset, - spa->spa_errlog_scrub, name, tx); + + if (!spa_feature_is_enabled(spa, SPA_FEATURE_HEAD_ERRLOG)) { + bookmark_to_name(&se->se_bookmark, name, sizeof (name)); + (void) zap_remove(spa->spa_meta_objset, + spa->spa_errlog_last, name, tx); + (void) zap_remove(spa->spa_meta_objset, + spa->spa_errlog_scrub, name, tx); + } else { + errphys_to_name(&se->se_zep, name, sizeof (name)); + zap_cursor_t zc; + zap_attribute_t za; + for (zap_cursor_init(&zc, spa->spa_meta_objset, + spa->spa_errlog_last); + zap_cursor_retrieve(&zc, &za) == 0; + zap_cursor_advance(&zc)) { + zap_remove(spa->spa_meta_objset, + za.za_first_integer, name, tx); + } + zap_cursor_fini(&zc); + + for (zap_cursor_init(&zc, spa->spa_meta_objset, + spa->spa_errlog_scrub); + zap_cursor_retrieve(&zc, &za) == 0; + zap_cursor_advance(&zc)) { + zap_remove(spa->spa_meta_objset, + za.za_first_integer, name, tx); + } + zap_cursor_fini(&zc); + } } } /* * Stash away healed bookmarks to remove them from the on-disk error logs * later in spa_remove_healed_errors(). */ void -spa_remove_error(spa_t *spa, zbookmark_phys_t *zb) +spa_remove_error(spa_t *spa, zbookmark_phys_t *zb, const uint64_t *birth) { - char name[NAME_MAX_LEN]; - - bookmark_to_name(zb, name, sizeof (name)); - - spa_add_healed_error(spa, spa->spa_errlog_last, zb); - spa_add_healed_error(spa, spa->spa_errlog_scrub, zb); + spa_add_healed_error(spa, spa->spa_errlog_last, zb, birth); + spa_add_healed_error(spa, spa->spa_errlog_scrub, zb, birth); } static uint64_t approx_errlog_size_impl(spa_t *spa, uint64_t spa_err_obj) { if (spa_err_obj == 0) return (0); uint64_t total = 0; zap_cursor_t zc; zap_attribute_t za; for (zap_cursor_init(&zc, spa->spa_meta_objset, spa_err_obj); zap_cursor_retrieve(&zc, &za) == 0; zap_cursor_advance(&zc)) { uint64_t count; if (zap_count(spa->spa_meta_objset, za.za_first_integer, &count) == 0) total += count; } zap_cursor_fini(&zc); return (total); } /* * Return the approximate number of errors currently in the error log. This * will be nonzero if there are some errors, but otherwise it may be more * or less than the number of entries returned by spa_get_errlog(). */ uint64_t spa_approx_errlog_size(spa_t *spa) { uint64_t total = 0; if (!spa_feature_is_enabled(spa, SPA_FEATURE_HEAD_ERRLOG)) { mutex_enter(&spa->spa_errlog_lock); uint64_t count; if (spa->spa_errlog_scrub != 0 && zap_count(spa->spa_meta_objset, spa->spa_errlog_scrub, &count) == 0) total += count; if (spa->spa_errlog_last != 0 && !spa->spa_scrub_finished && zap_count(spa->spa_meta_objset, spa->spa_errlog_last, &count) == 0) total += count; mutex_exit(&spa->spa_errlog_lock); } else { mutex_enter(&spa->spa_errlog_lock); total += approx_errlog_size_impl(spa, spa->spa_errlog_last); total += approx_errlog_size_impl(spa, spa->spa_errlog_scrub); mutex_exit(&spa->spa_errlog_lock); } mutex_enter(&spa->spa_errlist_lock); total += avl_numnodes(&spa->spa_errlist_last); total += avl_numnodes(&spa->spa_errlist_scrub); mutex_exit(&spa->spa_errlist_lock); return (total); } /* * This function sweeps through an on-disk error log and stores all bookmarks * as error bookmarks in a new ZAP object. At the end we discard the old one, * and spa_update_errlog() will set the spa's on-disk error log to new ZAP * object. */ static void sync_upgrade_errlog(spa_t *spa, uint64_t spa_err_obj, uint64_t *newobj, dmu_tx_t *tx) { zap_cursor_t zc; zap_attribute_t za; zbookmark_phys_t zb; uint64_t count; *newobj = zap_create(spa->spa_meta_objset, DMU_OT_ERROR_LOG, DMU_OT_NONE, 0, tx); /* * If we cannnot perform the upgrade we should clear the old on-disk * error logs. */ if (zap_count(spa->spa_meta_objset, spa_err_obj, &count) != 0) { VERIFY0(dmu_object_free(spa->spa_meta_objset, spa_err_obj, tx)); return; } for (zap_cursor_init(&zc, spa->spa_meta_objset, spa_err_obj); zap_cursor_retrieve(&zc, &za) == 0; zap_cursor_advance(&zc)) { if (spa_upgrade_errlog_limit != 0 && zc.zc_cd == spa_upgrade_errlog_limit) break; name_to_bookmark(za.za_name, &zb); zbookmark_err_phys_t zep; zep.zb_object = zb.zb_object; zep.zb_level = zb.zb_level; zep.zb_blkid = zb.zb_blkid; zep.zb_birth = 0; /* * In case of an error we should simply continue instead of * returning prematurely. See the next comment. */ uint64_t head_ds; dsl_pool_t *dp = spa->spa_dsl_pool; dsl_dataset_t *ds; objset_t *os; int error = dsl_dataset_hold_obj_flags(dp, zb.zb_objset, DS_HOLD_FLAG_DECRYPT, FTAG, &ds); if (error != 0) continue; head_ds = dsl_dir_phys(ds->ds_dir)->dd_head_dataset_obj; /* * The objset and the dnode are required for getting the block * pointer, which is used to determine if BP_IS_HOLE(). If * getting the objset or the dnode fails, do not create a * zap entry (presuming we know the dataset) as this may create * spurious errors that we cannot ever resolve. If an error is * truly persistent, it should re-appear after a scan. */ if (dmu_objset_from_ds(ds, &os) != 0) { dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); continue; } dnode_t *dn; blkptr_t bp; if (dnode_hold(os, zep.zb_object, FTAG, &dn) != 0) { dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); continue; } rw_enter(&dn->dn_struct_rwlock, RW_READER); error = dbuf_dnode_findbp(dn, zep.zb_level, zep.zb_blkid, &bp, NULL, NULL); if (error == EACCES) error = 0; else if (!error) zep.zb_birth = bp.blk_birth; rw_exit(&dn->dn_struct_rwlock); dnode_rele(dn, FTAG); dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); if (error != 0 || BP_IS_HOLE(&bp)) continue; uint64_t err_obj; error = zap_lookup_int_key(spa->spa_meta_objset, *newobj, head_ds, &err_obj); if (error == ENOENT) { err_obj = zap_create(spa->spa_meta_objset, DMU_OT_ERROR_LOG, DMU_OT_NONE, 0, tx); (void) zap_update_int_key(spa->spa_meta_objset, *newobj, head_ds, err_obj, tx); } char buf[64]; errphys_to_name(&zep, buf, sizeof (buf)); const char *name = ""; (void) zap_update(spa->spa_meta_objset, err_obj, buf, 1, strlen(name) + 1, name, tx); } zap_cursor_fini(&zc); VERIFY0(dmu_object_free(spa->spa_meta_objset, spa_err_obj, tx)); } void spa_upgrade_errlog(spa_t *spa, dmu_tx_t *tx) { uint64_t newobj = 0; mutex_enter(&spa->spa_errlog_lock); if (spa->spa_errlog_last != 0) { sync_upgrade_errlog(spa, spa->spa_errlog_last, &newobj, tx); spa->spa_errlog_last = newobj; } if (spa->spa_errlog_scrub != 0) { sync_upgrade_errlog(spa, spa->spa_errlog_scrub, &newobj, tx); spa->spa_errlog_scrub = newobj; } mutex_exit(&spa->spa_errlog_lock); } #ifdef _KERNEL /* * If an error block is shared by two datasets it will be counted twice. */ static int process_error_log(spa_t *spa, uint64_t obj, void *uaddr, uint64_t *count) { if (obj == 0) return (0); zap_cursor_t *zc; zap_attribute_t *za; zc = kmem_zalloc(sizeof (zap_cursor_t), KM_SLEEP); za = kmem_zalloc(sizeof (zap_attribute_t), KM_SLEEP); if (!spa_feature_is_enabled(spa, SPA_FEATURE_HEAD_ERRLOG)) { for (zap_cursor_init(zc, spa->spa_meta_objset, obj); zap_cursor_retrieve(zc, za) == 0; zap_cursor_advance(zc)) { if (*count == 0) { zap_cursor_fini(zc); kmem_free(zc, sizeof (*zc)); kmem_free(za, sizeof (*za)); return (SET_ERROR(ENOMEM)); } zbookmark_phys_t zb; name_to_bookmark(za->za_name, &zb); int error = copyout_entry(&zb, uaddr, count); if (error != 0) { zap_cursor_fini(zc); kmem_free(zc, sizeof (*zc)); kmem_free(za, sizeof (*za)); return (error); } } zap_cursor_fini(zc); kmem_free(zc, sizeof (*zc)); kmem_free(za, sizeof (*za)); return (0); } for (zap_cursor_init(zc, spa->spa_meta_objset, obj); zap_cursor_retrieve(zc, za) == 0; zap_cursor_advance(zc)) { zap_cursor_t *head_ds_cursor; zap_attribute_t *head_ds_attr; head_ds_cursor = kmem_zalloc(sizeof (zap_cursor_t), KM_SLEEP); head_ds_attr = kmem_zalloc(sizeof (zap_attribute_t), KM_SLEEP); uint64_t head_ds_err_obj = za->za_first_integer; uint64_t head_ds; name_to_object(za->za_name, &head_ds); for (zap_cursor_init(head_ds_cursor, spa->spa_meta_objset, head_ds_err_obj); zap_cursor_retrieve(head_ds_cursor, head_ds_attr) == 0; zap_cursor_advance(head_ds_cursor)) { zbookmark_err_phys_t head_ds_block; name_to_errphys(head_ds_attr->za_name, &head_ds_block); int error = process_error_block(spa, head_ds, &head_ds_block, uaddr, count); if (error != 0) { zap_cursor_fini(head_ds_cursor); kmem_free(head_ds_cursor, sizeof (*head_ds_cursor)); kmem_free(head_ds_attr, sizeof (*head_ds_attr)); zap_cursor_fini(zc); kmem_free(za, sizeof (*za)); kmem_free(zc, sizeof (*zc)); return (error); } } zap_cursor_fini(head_ds_cursor); kmem_free(head_ds_cursor, sizeof (*head_ds_cursor)); kmem_free(head_ds_attr, sizeof (*head_ds_attr)); } zap_cursor_fini(zc); kmem_free(za, sizeof (*za)); kmem_free(zc, sizeof (*zc)); return (0); } static int process_error_list(spa_t *spa, avl_tree_t *list, void *uaddr, uint64_t *count) { spa_error_entry_t *se; if (!spa_feature_is_enabled(spa, SPA_FEATURE_HEAD_ERRLOG)) { for (se = avl_first(list); se != NULL; se = AVL_NEXT(list, se)) { int error = copyout_entry(&se->se_bookmark, uaddr, count); if (error != 0) { return (error); } } return (0); } for (se = avl_first(list); se != NULL; se = AVL_NEXT(list, se)) { uint64_t head_ds = 0; int error = get_head_ds(spa, se->se_bookmark.zb_objset, &head_ds); /* * If get_head_ds() errors out, set the head filesystem * to the filesystem stored in the bookmark of the * error block. */ if (error != 0) head_ds = se->se_bookmark.zb_objset; error = process_error_block(spa, head_ds, &se->se_zep, uaddr, count); if (error != 0) return (error); } return (0); } #endif /* * Copy all known errors to userland as an array of bookmarks. This is * actually a union of the on-disk last log and current log, as well as any * pending error requests. * * Because the act of reading the on-disk log could cause errors to be * generated, we have two separate locks: one for the error log and one for the * in-core error lists. We only need the error list lock to log and error, so * we grab the error log lock while we read the on-disk logs, and only pick up * the error list lock when we are finished. */ int spa_get_errlog(spa_t *spa, void *uaddr, uint64_t *count) { int ret = 0; #ifdef _KERNEL /* * The pool config lock is needed to hold a dataset_t via (among other * places) process_error_list() -> process_error_block()-> * find_top_affected_fs(), and lock ordering requires that we get it * before the spa_errlog_lock. */ dsl_pool_config_enter(spa->spa_dsl_pool, FTAG); mutex_enter(&spa->spa_errlog_lock); ret = process_error_log(spa, spa->spa_errlog_scrub, uaddr, count); if (!ret && !spa->spa_scrub_finished) ret = process_error_log(spa, spa->spa_errlog_last, uaddr, count); mutex_enter(&spa->spa_errlist_lock); if (!ret) ret = process_error_list(spa, &spa->spa_errlist_scrub, uaddr, count); if (!ret) ret = process_error_list(spa, &spa->spa_errlist_last, uaddr, count); mutex_exit(&spa->spa_errlist_lock); mutex_exit(&spa->spa_errlog_lock); dsl_pool_config_exit(spa->spa_dsl_pool, FTAG); #else (void) spa, (void) uaddr, (void) count; #endif return (ret); } /* * Called when a scrub completes. This simply set a bit which tells which AVL * tree to add new errors. spa_errlog_sync() is responsible for actually * syncing the changes to the underlying objects. */ void spa_errlog_rotate(spa_t *spa) { mutex_enter(&spa->spa_errlist_lock); spa->spa_scrub_finished = B_TRUE; mutex_exit(&spa->spa_errlist_lock); } /* * Discard any pending errors from the spa_t. Called when unloading a faulted * pool, as the errors encountered during the open cannot be synced to disk. */ void spa_errlog_drain(spa_t *spa) { spa_error_entry_t *se; void *cookie; mutex_enter(&spa->spa_errlist_lock); cookie = NULL; while ((se = avl_destroy_nodes(&spa->spa_errlist_last, &cookie)) != NULL) kmem_free(se, sizeof (spa_error_entry_t)); cookie = NULL; while ((se = avl_destroy_nodes(&spa->spa_errlist_scrub, &cookie)) != NULL) kmem_free(se, sizeof (spa_error_entry_t)); mutex_exit(&spa->spa_errlist_lock); } /* * Process a list of errors into the current on-disk log. */ void sync_error_list(spa_t *spa, avl_tree_t *t, uint64_t *obj, dmu_tx_t *tx) { spa_error_entry_t *se; char buf[NAME_MAX_LEN]; void *cookie; if (avl_numnodes(t) == 0) return; /* create log if necessary */ if (*obj == 0) *obj = zap_create(spa->spa_meta_objset, DMU_OT_ERROR_LOG, DMU_OT_NONE, 0, tx); /* add errors to the current log */ if (!spa_feature_is_enabled(spa, SPA_FEATURE_HEAD_ERRLOG)) { for (se = avl_first(t); se != NULL; se = AVL_NEXT(t, se)) { bookmark_to_name(&se->se_bookmark, buf, sizeof (buf)); const char *name = se->se_name ? se->se_name : ""; (void) zap_update(spa->spa_meta_objset, *obj, buf, 1, strlen(name) + 1, name, tx); } } else { for (se = avl_first(t); se != NULL; se = AVL_NEXT(t, se)) { zbookmark_err_phys_t zep; zep.zb_object = se->se_zep.zb_object; zep.zb_level = se->se_zep.zb_level; zep.zb_blkid = se->se_zep.zb_blkid; zep.zb_birth = se->se_zep.zb_birth; uint64_t head_ds = 0; int error = get_head_ds(spa, se->se_bookmark.zb_objset, &head_ds); /* * If get_head_ds() errors out, set the head filesystem * to the filesystem stored in the bookmark of the * error block. */ if (error != 0) head_ds = se->se_bookmark.zb_objset; uint64_t err_obj; error = zap_lookup_int_key(spa->spa_meta_objset, *obj, head_ds, &err_obj); if (error == ENOENT) { err_obj = zap_create(spa->spa_meta_objset, DMU_OT_ERROR_LOG, DMU_OT_NONE, 0, tx); (void) zap_update_int_key(spa->spa_meta_objset, *obj, head_ds, err_obj, tx); } errphys_to_name(&zep, buf, sizeof (buf)); const char *name = se->se_name ? se->se_name : ""; (void) zap_update(spa->spa_meta_objset, err_obj, buf, 1, strlen(name) + 1, name, tx); } } /* purge the error list */ cookie = NULL; while ((se = avl_destroy_nodes(t, &cookie)) != NULL) kmem_free(se, sizeof (spa_error_entry_t)); } static void delete_errlog(spa_t *spa, uint64_t spa_err_obj, dmu_tx_t *tx) { if (spa_feature_is_enabled(spa, SPA_FEATURE_HEAD_ERRLOG)) { zap_cursor_t zc; zap_attribute_t za; for (zap_cursor_init(&zc, spa->spa_meta_objset, spa_err_obj); zap_cursor_retrieve(&zc, &za) == 0; zap_cursor_advance(&zc)) { VERIFY0(dmu_object_free(spa->spa_meta_objset, za.za_first_integer, tx)); } zap_cursor_fini(&zc); } VERIFY0(dmu_object_free(spa->spa_meta_objset, spa_err_obj, tx)); } /* * Sync the error log out to disk. This is a little tricky because the act of * writing the error log requires the spa_errlist_lock. So, we need to lock the * error lists, take a copy of the lists, and then reinitialize them. Then, we * drop the error list lock and take the error log lock, at which point we * do the errlog processing. Then, if we encounter an I/O error during this * process, we can successfully add the error to the list. Note that this will * result in the perpetual recycling of errors, but it is an unlikely situation * and not a performance critical operation. */ void spa_errlog_sync(spa_t *spa, uint64_t txg) { dmu_tx_t *tx; avl_tree_t scrub, last; int scrub_finished; mutex_enter(&spa->spa_errlist_lock); /* * Bail out early under normal circumstances. */ if (avl_numnodes(&spa->spa_errlist_scrub) == 0 && avl_numnodes(&spa->spa_errlist_last) == 0 && avl_numnodes(&spa->spa_errlist_healed) == 0 && !spa->spa_scrub_finished) { mutex_exit(&spa->spa_errlist_lock); return; } spa_get_errlists(spa, &last, &scrub); scrub_finished = spa->spa_scrub_finished; spa->spa_scrub_finished = B_FALSE; mutex_exit(&spa->spa_errlist_lock); /* * The pool config lock is needed to hold a dataset_t via * sync_error_list() -> get_head_ds(), and lock ordering * requires that we get it before the spa_errlog_lock. */ dsl_pool_config_enter(spa->spa_dsl_pool, FTAG); mutex_enter(&spa->spa_errlog_lock); tx = dmu_tx_create_assigned(spa->spa_dsl_pool, txg); /* * Remove healed errors from errors. */ spa_remove_healed_errors(spa, &last, &scrub, tx); /* * Sync out the current list of errors. */ sync_error_list(spa, &last, &spa->spa_errlog_last, tx); /* * Rotate the log if necessary. */ if (scrub_finished) { if (spa->spa_errlog_last != 0) delete_errlog(spa, spa->spa_errlog_last, tx); spa->spa_errlog_last = spa->spa_errlog_scrub; spa->spa_errlog_scrub = 0; sync_error_list(spa, &scrub, &spa->spa_errlog_last, tx); } /* * Sync out any pending scrub errors. */ sync_error_list(spa, &scrub, &spa->spa_errlog_scrub, tx); /* * Update the MOS to reflect the new values. */ (void) zap_update(spa->spa_meta_objset, DMU_POOL_DIRECTORY_OBJECT, DMU_POOL_ERRLOG_LAST, sizeof (uint64_t), 1, &spa->spa_errlog_last, tx); (void) zap_update(spa->spa_meta_objset, DMU_POOL_DIRECTORY_OBJECT, DMU_POOL_ERRLOG_SCRUB, sizeof (uint64_t), 1, &spa->spa_errlog_scrub, tx); dmu_tx_commit(tx); mutex_exit(&spa->spa_errlog_lock); dsl_pool_config_exit(spa->spa_dsl_pool, FTAG); } static void delete_dataset_errlog(spa_t *spa, uint64_t spa_err_obj, uint64_t ds, dmu_tx_t *tx) { if (spa_err_obj == 0) return; zap_cursor_t zc; zap_attribute_t za; for (zap_cursor_init(&zc, spa->spa_meta_objset, spa_err_obj); zap_cursor_retrieve(&zc, &za) == 0; zap_cursor_advance(&zc)) { uint64_t head_ds; name_to_object(za.za_name, &head_ds); if (head_ds == ds) { (void) zap_remove(spa->spa_meta_objset, spa_err_obj, za.za_name, tx); VERIFY0(dmu_object_free(spa->spa_meta_objset, za.za_first_integer, tx)); break; } } zap_cursor_fini(&zc); } void spa_delete_dataset_errlog(spa_t *spa, uint64_t ds, dmu_tx_t *tx) { mutex_enter(&spa->spa_errlog_lock); delete_dataset_errlog(spa, spa->spa_errlog_scrub, ds, tx); delete_dataset_errlog(spa, spa->spa_errlog_last, ds, tx); mutex_exit(&spa->spa_errlog_lock); } static int find_txg_ancestor_snapshot(spa_t *spa, uint64_t new_head, uint64_t old_head, uint64_t *txg) { dsl_dataset_t *ds; dsl_pool_t *dp = spa->spa_dsl_pool; int error = dsl_dataset_hold_obj_flags(dp, old_head, DS_HOLD_FLAG_DECRYPT, FTAG, &ds); if (error != 0) return (error); uint64_t prev_obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; uint64_t prev_obj_txg = dsl_dataset_phys(ds)->ds_prev_snap_txg; while (prev_obj != 0) { dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); if ((error = dsl_dataset_hold_obj_flags(dp, prev_obj, DS_HOLD_FLAG_DECRYPT, FTAG, &ds)) == 0 && dsl_dir_phys(ds->ds_dir)->dd_head_dataset_obj == new_head) break; if (error != 0) return (error); prev_obj_txg = dsl_dataset_phys(ds)->ds_prev_snap_txg; prev_obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; } dsl_dataset_rele_flags(ds, DS_HOLD_FLAG_DECRYPT, FTAG); ASSERT(prev_obj != 0); *txg = prev_obj_txg; return (0); } static void swap_errlog(spa_t *spa, uint64_t spa_err_obj, uint64_t new_head, uint64_t old_head, dmu_tx_t *tx) { if (spa_err_obj == 0) return; uint64_t old_head_errlog; int error = zap_lookup_int_key(spa->spa_meta_objset, spa_err_obj, old_head, &old_head_errlog); /* If no error log, then there is nothing to do. */ if (error != 0) return; uint64_t txg; error = find_txg_ancestor_snapshot(spa, new_head, old_head, &txg); if (error != 0) return; /* * Create an error log if the file system being promoted does not * already have one. */ uint64_t new_head_errlog; error = zap_lookup_int_key(spa->spa_meta_objset, spa_err_obj, new_head, &new_head_errlog); if (error != 0) { new_head_errlog = zap_create(spa->spa_meta_objset, DMU_OT_ERROR_LOG, DMU_OT_NONE, 0, tx); (void) zap_update_int_key(spa->spa_meta_objset, spa_err_obj, new_head, new_head_errlog, tx); } zap_cursor_t zc; zap_attribute_t za; zbookmark_err_phys_t err_block; for (zap_cursor_init(&zc, spa->spa_meta_objset, old_head_errlog); zap_cursor_retrieve(&zc, &za) == 0; zap_cursor_advance(&zc)) { const char *name = ""; name_to_errphys(za.za_name, &err_block); if (err_block.zb_birth < txg) { (void) zap_update(spa->spa_meta_objset, new_head_errlog, za.za_name, 1, strlen(name) + 1, name, tx); (void) zap_remove(spa->spa_meta_objset, old_head_errlog, za.za_name, tx); } } zap_cursor_fini(&zc); } void spa_swap_errlog(spa_t *spa, uint64_t new_head_ds, uint64_t old_head_ds, dmu_tx_t *tx) { mutex_enter(&spa->spa_errlog_lock); swap_errlog(spa, spa->spa_errlog_scrub, new_head_ds, old_head_ds, tx); swap_errlog(spa, spa->spa_errlog_last, new_head_ds, old_head_ds, tx); mutex_exit(&spa->spa_errlog_lock); } #if defined(_KERNEL) /* error handling */ EXPORT_SYMBOL(spa_log_error); EXPORT_SYMBOL(spa_approx_errlog_size); EXPORT_SYMBOL(spa_get_errlog); EXPORT_SYMBOL(spa_errlog_rotate); EXPORT_SYMBOL(spa_errlog_drain); EXPORT_SYMBOL(spa_errlog_sync); EXPORT_SYMBOL(spa_get_errlists); EXPORT_SYMBOL(spa_delete_dataset_errlog); EXPORT_SYMBOL(spa_swap_errlog); EXPORT_SYMBOL(sync_error_list); EXPORT_SYMBOL(spa_upgrade_errlog); #endif /* BEGIN CSTYLED */ ZFS_MODULE_PARAM(zfs_spa, spa_, upgrade_errlog_limit, UINT, ZMOD_RW, "Limit the number of errors which will be upgraded to the new " "on-disk error log when enabling head_errlog"); /* END CSTYLED */ diff --git a/tests/zfs-tests/tests/functional/cli_root/zpool_status/zpool_status_007_pos.ksh b/tests/zfs-tests/tests/functional/cli_root/zpool_status/zpool_status_007_pos.ksh index c9849379f779..666ac9bfc9dd 100755 --- a/tests/zfs-tests/tests/functional/cli_root/zpool_status/zpool_status_007_pos.ksh +++ b/tests/zfs-tests/tests/functional/cli_root/zpool_status/zpool_status_007_pos.ksh @@ -1,98 +1,111 @@ #!/bin/ksh -p # # 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 https://opensource.org/licenses/CDDL-1.0. # 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) 2023 George Amanakis. All rights reserved. # # # DESCRIPTION: # Verify reporting errors when deleting corrupted files after scrub # # STRATEGY: # 1. Create a pool, and a file # 2. Corrupt the file # 3. Create snapshots and clones like: # fs->snap1->clone1->snap2->clone2->... # 4. Read the original file and immediately delete it # 5. Delete the file in clone2 # 6. Snapshot clone2->snapxx and clone into snapxx->clonexx # 7. Verify we report errors in the pool in 'zpool status -v' # 8. Promote clone1 # 9. Verify we report errors in the pool in 'zpool status -v' +# 10. Delete the corrupted file and origin snapshots. +# 11. Verify we do not report data errors anymore, without requiring +# a scrub. . $STF_SUITE/include/libtest.shlib verify_runnable "both" function cleanup { destroy_pool $TESTPOOL2 rm -f $TESTDIR/vdev_a } log_assert "Verify reporting errors when deleting corrupted files after scrub" log_onexit cleanup typeset file="/$TESTPOOL2/$TESTFS1/$TESTFILE0" truncate -s $MINVDEVSIZE $TESTDIR/vdev_a log_must zpool create -f $TESTPOOL2 $TESTDIR/vdev_a log_must zfs create -o primarycache=none $TESTPOOL2/$TESTFS1 log_must dd if=/dev/urandom of=$file bs=1024 count=1024 oflag=sync corrupt_blocks_at_level $file 0 lastfs="$(zfs list -r $TESTPOOL2 | tail -1 | awk '{print $1}')" for i in {1..3}; do log_must zfs snap $lastfs@snap$i log_must zfs clone $lastfs@snap$i $TESTPOOL2/clone$i lastfs="$(zfs list -r $TESTPOOL2/clone$i | tail -1 | awk '{print $1}')" done log_must zpool scrub -w $TESTPOOL2 log_must rm $file /$TESTPOOL2/clone2/$TESTFILE0 log_must zfs snap $TESTPOOL2/clone2@snapxx log_must zfs clone $TESTPOOL2/clone2@snapxx $TESTPOOL2/clonexx log_must zpool status -v $TESTPOOL2 log_must eval "zpool status -v $TESTPOOL2 | \ grep \"Permanent errors have been detected\"" log_must eval "zpool status -v | grep '$TESTPOOL2/$TESTFS1@snap1:/$TESTFILE0'" log_must eval "zpool status -v | grep '$TESTPOOL2/clone1/$TESTFILE0'" log_must eval "zpool status -v | grep '$TESTPOOL2/clone1@snap2:/$TESTFILE0'" log_mustnot eval "zpool status -v | grep '$TESTPOOL2/clone2/$TESTFILE0'" log_mustnot eval "zpool status -v | grep '$TESTPOOL2/clonexx/$TESTFILE0'" log_must eval "zpool status -v | grep '$TESTPOOL2/clone2@snap3:/$TESTFILE0'" log_must eval "zpool status -v | grep '$TESTPOOL2/clone3/$TESTFILE0'" log_must zfs promote $TESTPOOL2/clone1 log_must eval "zpool status -v $TESTPOOL2 | \ grep \"Permanent errors have been detected\"" log_must eval "zpool status -v | grep '$TESTPOOL2/clone1@snap1:/$TESTFILE0'" log_must eval "zpool status -v | grep '$TESTPOOL2/clone1/$TESTFILE0'" log_must eval "zpool status -v | grep '$TESTPOOL2/clone1@snap2:/$TESTFILE0'" log_mustnot eval "zpool status -v | grep '$TESTPOOL2/clone2/$TESTFILE0'" log_mustnot eval "zpool status -v | grep '$TESTPOOL2/clonexx/$TESTFILE0'" log_must eval "zpool status -v | grep '$TESTPOOL2/clone2@snap3:/$TESTFILE0'" log_must eval "zpool status -v | grep '$TESTPOOL2/clone3/$TESTFILE0'" +log_must rm /$TESTPOOL2/clone1/$TESTFILE0 +log_must zfs destroy -R $TESTPOOL2/clone1@snap1 +log_must zfs destroy -R $TESTPOOL2/clone1@snap2 +log_must zfs list -r $TESTPOOL2 +log_must zpool status -v $TESTPOOL2 +log_must zpool sync +log_must zpool status -v $TESTPOOL2 +log_must eval "zpool status -v $TESTPOOL2 | \ + grep \"No known data errors\"" + log_pass "Verify reporting errors when deleting corrupted files after scrub"