diff --git a/cmd/zstream/zstream_decompress.c b/cmd/zstream/zstream_decompress.c index f5f66080d060..f8f439d4626d 100644 --- a/cmd/zstream/zstream_decompress.c +++ b/cmd/zstream/zstream_decompress.c @@ -1,386 +1,388 @@ /* * 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 2022 Axcient. All rights reserved. * Use is subject to license terms. * * Copyright (c) 2024, Klara, Inc. */ #include #include #include #include #include #include #include #include #include "zfs_fletcher.h" #include "zstream.h" static int dump_record(dmu_replay_record_t *drr, void *payload, int payload_len, zio_cksum_t *zc, int outfd) { assert(offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum) == sizeof (dmu_replay_record_t) - sizeof (zio_cksum_t)); fletcher_4_incremental_native(drr, offsetof(dmu_replay_record_t, drr_u.drr_checksum.drr_checksum), zc); if (drr->drr_type != DRR_BEGIN) { assert(ZIO_CHECKSUM_IS_ZERO(&drr->drr_u. drr_checksum.drr_checksum)); drr->drr_u.drr_checksum.drr_checksum = *zc; } fletcher_4_incremental_native(&drr->drr_u.drr_checksum.drr_checksum, sizeof (zio_cksum_t), zc); if (write(outfd, drr, sizeof (*drr)) == -1) return (errno); if (payload_len != 0) { fletcher_4_incremental_native(payload, payload_len, zc); if (write(outfd, payload, payload_len) == -1) return (errno); } return (0); } int zstream_do_decompress(int argc, char *argv[]) { const int KEYSIZE = 64; int bufsz = SPA_MAXBLOCKSIZE; char *buf = safe_malloc(bufsz); dmu_replay_record_t thedrr; dmu_replay_record_t *drr = &thedrr; zio_cksum_t stream_cksum; int c; boolean_t verbose = B_FALSE; while ((c = getopt(argc, argv, "v")) != -1) { switch (c) { case 'v': verbose = B_TRUE; break; case '?': (void) fprintf(stderr, "invalid option '%c'\n", optopt); zstream_usage(); break; } } argc -= optind; argv += optind; if (argc < 0) zstream_usage(); if (hcreate(argc) == 0) errx(1, "hcreate"); for (int i = 0; i < argc; i++) { uint64_t object, offset; char *obj_str; char *offset_str; char *key; char *end; enum zio_compress type = ZIO_COMPRESS_LZ4; obj_str = strsep(&argv[i], ","); if (argv[i] == NULL) { zstream_usage(); exit(2); } errno = 0; object = strtoull(obj_str, &end, 0); if (errno || *end != '\0') errx(1, "invalid value for object"); offset_str = strsep(&argv[i], ","); offset = strtoull(offset_str, &end, 0); if (errno || *end != '\0') errx(1, "invalid value for offset"); if (argv[i]) { if (0 == strcmp("off", argv[i])) type = ZIO_COMPRESS_OFF; else if (0 == strcmp("lz4", argv[i])) type = ZIO_COMPRESS_LZ4; else if (0 == strcmp("lzjb", argv[i])) type = ZIO_COMPRESS_LZJB; else if (0 == strcmp("gzip", argv[i])) type = ZIO_COMPRESS_GZIP_1; else if (0 == strcmp("zle", argv[i])) type = ZIO_COMPRESS_ZLE; else if (0 == strcmp("zstd", argv[i])) type = ZIO_COMPRESS_ZSTD; else { fprintf(stderr, "Invalid compression type %s.\n" "Supported types are off, lz4, lzjb, gzip, " "zle, and zstd\n", argv[i]); exit(2); } } if (asprintf(&key, "%llu,%llu", (u_longlong_t)object, (u_longlong_t)offset) < 0) { err(1, "asprintf"); } ENTRY e = {.key = key}; ENTRY *p; p = hsearch(e, ENTER); if (p == NULL) errx(1, "hsearch"); p->data = (void*)(intptr_t)type; } if (isatty(STDIN_FILENO)) { (void) fprintf(stderr, "Error: The send stream is a binary format " "and can not be read from a\n" "terminal. Standard input must be redirected.\n"); exit(1); } fletcher_4_init(); int begin = 0; boolean_t seen = B_FALSE; while (sfread(drr, sizeof (*drr), stdin) != 0) { struct drr_write *drrw; uint64_t payload_size = 0; /* * We need to regenerate the checksum. */ if (drr->drr_type != DRR_BEGIN) { memset(&drr->drr_u.drr_checksum.drr_checksum, 0, sizeof (drr->drr_u.drr_checksum.drr_checksum)); } switch (drr->drr_type) { case DRR_BEGIN: { ZIO_SET_CHECKSUM(&stream_cksum, 0, 0, 0, 0); VERIFY0(begin++); seen = B_TRUE; uint32_t sz = drr->drr_payloadlen; VERIFY3U(sz, <=, 1U << 28); if (sz != 0) { if (sz > bufsz) { buf = realloc(buf, sz); if (buf == NULL) err(1, "realloc"); bufsz = sz; } (void) sfread(buf, sz, stdin); } payload_size = sz; break; } case DRR_END: { struct drr_end *drre = &drr->drr_u.drr_end; /* * We would prefer to just check --begin == 0, but * replication streams have an end of stream END * record, so we must avoid tripping it. */ VERIFY3B(seen, ==, B_TRUE); begin--; /* * Use the recalculated checksum, unless this is * the END record of a stream package, which has * no checksum. */ if (!ZIO_CHECKSUM_IS_ZERO(&drre->drr_checksum)) drre->drr_checksum = stream_cksum; break; } case DRR_OBJECT: { struct drr_object *drro = &drr->drr_u.drr_object; VERIFY3S(begin, ==, 1); if (drro->drr_bonuslen > 0) { payload_size = DRR_OBJECT_PAYLOAD_SIZE(drro); (void) sfread(buf, payload_size, stdin); } break; } case DRR_SPILL: { struct drr_spill *drrs = &drr->drr_u.drr_spill; VERIFY3S(begin, ==, 1); payload_size = DRR_SPILL_PAYLOAD_SIZE(drrs); (void) sfread(buf, payload_size, stdin); break; } case DRR_WRITE_BYREF: VERIFY3S(begin, ==, 1); fprintf(stderr, "Deduplicated streams are not supported\n"); exit(1); break; case DRR_WRITE: { VERIFY3S(begin, ==, 1); drrw = &thedrr.drr_u.drr_write; payload_size = DRR_WRITE_PAYLOAD_SIZE(drrw); ENTRY *p; char key[KEYSIZE]; snprintf(key, KEYSIZE, "%llu,%llu", (u_longlong_t)drrw->drr_object, (u_longlong_t)drrw->drr_offset); ENTRY e = {.key = key}; p = hsearch(e, FIND); if (p == NULL) { /* * Read the contents of the block unaltered */ (void) sfread(buf, payload_size, stdin); break; } /* * Read and decompress the block */ enum zio_compress c = (enum zio_compress)(intptr_t)p->data; if (c == ZIO_COMPRESS_OFF) { (void) sfread(buf, payload_size, stdin); drrw->drr_compressiontype = ZIO_COMPRESS_OFF; if (verbose) fprintf(stderr, "Resetting compression type to " "off for ino %llu offset %llu\n", (u_longlong_t)drrw->drr_object, (u_longlong_t)drrw->drr_offset); break; } char *lzbuf = safe_calloc(payload_size); (void) sfread(lzbuf, payload_size, stdin); - abd_t sabd; + abd_t sabd, dabd; abd_get_from_buf_struct(&sabd, lzbuf, payload_size); - int err = zio_decompress_data(c, &sabd, buf, + abd_get_from_buf_struct(&dabd, buf, payload_size); + int err = zio_decompress_data(c, &sabd, &dabd, payload_size, payload_size, NULL); + abd_free(&dabd); abd_free(&sabd); if (err != 0) { /* * The block must not be compressed, at least * not with this compression type, possibly * because it gets written multiple times in * this stream. */ warnx("decompression failed for " "ino %llu offset %llu", (u_longlong_t)drrw->drr_object, (u_longlong_t)drrw->drr_offset); memcpy(buf, lzbuf, payload_size); } else if (verbose) { drrw->drr_compressiontype = ZIO_COMPRESS_OFF; fprintf(stderr, "successfully decompressed " "ino %llu offset %llu\n", (u_longlong_t)drrw->drr_object, (u_longlong_t)drrw->drr_offset); } else { drrw->drr_compressiontype = ZIO_COMPRESS_OFF; } free(lzbuf); break; } case DRR_WRITE_EMBEDDED: { VERIFY3S(begin, ==, 1); struct drr_write_embedded *drrwe = &drr->drr_u.drr_write_embedded; payload_size = P2ROUNDUP((uint64_t)drrwe->drr_psize, 8); (void) sfread(buf, payload_size, stdin); break; } case DRR_FREEOBJECTS: case DRR_FREE: case DRR_OBJECT_RANGE: VERIFY3S(begin, ==, 1); break; default: (void) fprintf(stderr, "INVALID record type 0x%x\n", drr->drr_type); /* should never happen, so assert */ assert(B_FALSE); } if (feof(stdout)) { fprintf(stderr, "Error: unexpected end-of-file\n"); exit(1); } if (ferror(stdout)) { fprintf(stderr, "Error while reading file: %s\n", strerror(errno)); exit(1); } /* * We need to recalculate the checksum, and it needs to be * initially zero to do that. BEGIN records don't have * a checksum. */ if (drr->drr_type != DRR_BEGIN) { memset(&drr->drr_u.drr_checksum.drr_checksum, 0, sizeof (drr->drr_u.drr_checksum.drr_checksum)); } if (dump_record(drr, buf, payload_size, &stream_cksum, STDOUT_FILENO) != 0) break; if (drr->drr_type == DRR_END) { /* * Typically the END record is either the last * thing in the stream, or it is followed * by a BEGIN record (which also zeros the checksum). * However, a stream package ends with two END * records. The last END record's checksum starts * from zero. */ ZIO_SET_CHECKSUM(&stream_cksum, 0, 0, 0, 0); } } free(buf); fletcher_4_fini(); hdestroy(); return (0); } diff --git a/include/sys/zio_compress.h b/include/sys/zio_compress.h index 56376fdd10a0..d41b5dfd447f 100644 --- a/include/sys/zio_compress.h +++ b/include/sys/zio_compress.h @@ -1,186 +1,220 @@ /* * 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 2009 Sun Microsystems, Inc. All rights reserved. * Copyright (c) 2019, Allan Jude - * Copyright (c) 2019, Klara Inc. + * Copyright (c) 2019, 2024, Klara, Inc. * Use is subject to license terms. * Copyright (c) 2015, 2016 by Delphix. All rights reserved. */ #ifndef _SYS_ZIO_COMPRESS_H #define _SYS_ZIO_COMPRESS_H #include #ifdef __cplusplus extern "C" { #endif enum zio_compress { ZIO_COMPRESS_INHERIT = 0, ZIO_COMPRESS_ON, ZIO_COMPRESS_OFF, ZIO_COMPRESS_LZJB, ZIO_COMPRESS_EMPTY, ZIO_COMPRESS_GZIP_1, ZIO_COMPRESS_GZIP_2, ZIO_COMPRESS_GZIP_3, ZIO_COMPRESS_GZIP_4, ZIO_COMPRESS_GZIP_5, ZIO_COMPRESS_GZIP_6, ZIO_COMPRESS_GZIP_7, ZIO_COMPRESS_GZIP_8, ZIO_COMPRESS_GZIP_9, ZIO_COMPRESS_ZLE, ZIO_COMPRESS_LZ4, ZIO_COMPRESS_ZSTD, ZIO_COMPRESS_FUNCTIONS }; /* Compression algorithms that have levels */ #define ZIO_COMPRESS_HASLEVEL(compress) ((compress == ZIO_COMPRESS_ZSTD || \ (compress >= ZIO_COMPRESS_GZIP_1 && \ compress <= ZIO_COMPRESS_GZIP_9))) #define ZIO_COMPLEVEL_INHERIT 0 #define ZIO_COMPLEVEL_DEFAULT 255 enum zio_zstd_levels { ZIO_ZSTD_LEVEL_INHERIT = 0, ZIO_ZSTD_LEVEL_1, #define ZIO_ZSTD_LEVEL_MIN ZIO_ZSTD_LEVEL_1 ZIO_ZSTD_LEVEL_2, ZIO_ZSTD_LEVEL_3, #define ZIO_ZSTD_LEVEL_DEFAULT ZIO_ZSTD_LEVEL_3 ZIO_ZSTD_LEVEL_4, ZIO_ZSTD_LEVEL_5, ZIO_ZSTD_LEVEL_6, ZIO_ZSTD_LEVEL_7, ZIO_ZSTD_LEVEL_8, ZIO_ZSTD_LEVEL_9, ZIO_ZSTD_LEVEL_10, ZIO_ZSTD_LEVEL_11, ZIO_ZSTD_LEVEL_12, ZIO_ZSTD_LEVEL_13, ZIO_ZSTD_LEVEL_14, ZIO_ZSTD_LEVEL_15, ZIO_ZSTD_LEVEL_16, ZIO_ZSTD_LEVEL_17, ZIO_ZSTD_LEVEL_18, ZIO_ZSTD_LEVEL_19, #define ZIO_ZSTD_LEVEL_MAX ZIO_ZSTD_LEVEL_19 ZIO_ZSTD_LEVEL_RESERVE = 101, /* Leave room for new positive levels */ ZIO_ZSTD_LEVEL_FAST, /* Fast levels are negative */ ZIO_ZSTD_LEVEL_FAST_1, #define ZIO_ZSTD_LEVEL_FAST_DEFAULT ZIO_ZSTD_LEVEL_FAST_1 ZIO_ZSTD_LEVEL_FAST_2, ZIO_ZSTD_LEVEL_FAST_3, ZIO_ZSTD_LEVEL_FAST_4, ZIO_ZSTD_LEVEL_FAST_5, ZIO_ZSTD_LEVEL_FAST_6, ZIO_ZSTD_LEVEL_FAST_7, ZIO_ZSTD_LEVEL_FAST_8, ZIO_ZSTD_LEVEL_FAST_9, ZIO_ZSTD_LEVEL_FAST_10, ZIO_ZSTD_LEVEL_FAST_20, ZIO_ZSTD_LEVEL_FAST_30, ZIO_ZSTD_LEVEL_FAST_40, ZIO_ZSTD_LEVEL_FAST_50, ZIO_ZSTD_LEVEL_FAST_60, ZIO_ZSTD_LEVEL_FAST_70, ZIO_ZSTD_LEVEL_FAST_80, ZIO_ZSTD_LEVEL_FAST_90, ZIO_ZSTD_LEVEL_FAST_100, ZIO_ZSTD_LEVEL_FAST_500, ZIO_ZSTD_LEVEL_FAST_1000, #define ZIO_ZSTD_LEVEL_FAST_MAX ZIO_ZSTD_LEVEL_FAST_1000 ZIO_ZSTD_LEVEL_AUTO = 251, /* Reserved for future use */ ZIO_ZSTD_LEVEL_LEVELS }; /* Forward Declaration to avoid visibility problems */ struct zio_prop; /* Common signature for all zio compress functions. */ -typedef size_t zio_compress_func_t(void *src, void *dst, +typedef size_t zio_compress_func_t(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int); /* Common signature for all zio decompress functions. */ -typedef int zio_decompress_func_t(void *src, void *dst, +typedef int zio_decompress_func_t(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int); /* Common signature for all zio decompress and get level functions. */ -typedef int zio_decompresslevel_func_t(void *src, void *dst, +typedef int zio_decompresslevel_func_t(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, uint8_t *level); /* * Information about each compression function. */ typedef const struct zio_compress_info { const char *ci_name; int ci_level; zio_compress_func_t *ci_compress; zio_decompress_func_t *ci_decompress; zio_decompresslevel_func_t *ci_decompress_level; } zio_compress_info_t; extern zio_compress_info_t zio_compress_table[ZIO_COMPRESS_FUNCTIONS]; /* * lz4 compression init & free */ extern void lz4_init(void); extern void lz4_fini(void); /* * Compression routines. */ -extern size_t zfs_lzjb_compress(void *src, void *dst, size_t s_len, +extern size_t zfs_lzjb_compress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int level); -extern int zfs_lzjb_decompress(void *src, void *dst, size_t s_len, +extern int zfs_lzjb_decompress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int level); -extern size_t zfs_gzip_compress(void *src, void *dst, size_t s_len, +extern size_t zfs_gzip_compress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int level); -extern int zfs_gzip_decompress(void *src, void *dst, size_t s_len, +extern int zfs_gzip_decompress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int level); -extern size_t zfs_zle_compress(void *src, void *dst, size_t s_len, +extern size_t zfs_zle_compress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int level); -extern int zfs_zle_decompress(void *src, void *dst, size_t s_len, +extern int zfs_zle_decompress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int level); -extern size_t zfs_lz4_compress(void *src, void *dst, size_t s_len, +extern size_t zfs_lz4_compress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int level); -extern int zfs_lz4_decompress(void *src, void *dst, size_t s_len, +extern int zfs_lz4_decompress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int level); /* * Compress and decompress data if necessary. */ extern size_t zio_compress_data(enum zio_compress c, abd_t *src, void **dst, size_t s_len, uint8_t level); extern int zio_decompress_data(enum zio_compress c, abd_t *src, void *dst, size_t s_len, size_t d_len, uint8_t *level); extern int zio_compress_to_feature(enum zio_compress comp); +#define ZFS_COMPRESS_WRAP_DECL(name) \ +size_t \ +name(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int n) \ +{ \ + void *s_buf = abd_borrow_buf_copy(src, s_len); \ + void *d_buf = abd_borrow_buf(dst, d_len); \ + size_t c_len = name##_buf(s_buf, d_buf, s_len, d_len, n); \ + abd_return_buf(src, s_buf, s_len); \ + abd_return_buf_copy(dst, d_buf, d_len); \ + return (c_len); \ +} +#define ZFS_DECOMPRESS_WRAP_DECL(name) \ +int \ +name(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int n) \ +{ \ + void *s_buf = abd_borrow_buf_copy(src, s_len); \ + void *d_buf = abd_borrow_buf(dst, d_len); \ + int err = name##_buf(s_buf, d_buf, s_len, d_len, n); \ + abd_return_buf(src, s_buf, s_len); \ + abd_return_buf_copy(dst, d_buf, d_len); \ + return (err); \ +} +#define ZFS_DECOMPRESS_LEVEL_WRAP_DECL(name) \ +int \ +name(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, uint8_t *n) \ +{ \ + void *s_buf = abd_borrow_buf_copy(src, s_len); \ + void *d_buf = abd_borrow_buf(dst, d_len); \ + int err = name##_buf(s_buf, d_buf, s_len, d_len, n); \ + abd_return_buf(src, s_buf, s_len); \ + abd_return_buf_copy(dst, d_buf, d_len); \ + return (err); \ +} + #ifdef __cplusplus } #endif #endif /* _SYS_ZIO_COMPRESS_H */ diff --git a/include/sys/zstd/zstd.h b/include/sys/zstd/zstd.h index f9e7ac0b32ea..6d212b082f9a 100644 --- a/include/sys/zstd/zstd.h +++ b/include/sys/zstd/zstd.h @@ -1,231 +1,231 @@ /* * BSD 3-Clause New License (https://spdx.org/licenses/BSD-3-Clause.html) * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: * * 1. Redistributions of source code must retain the above copyright notice, * this list of conditions and the following disclaimer. * * 2. Redistributions in binary form must reproduce the above copyright notice, * this list of conditions and the following disclaimer in the documentation * and/or other materials provided with the distribution. * * 3. Neither the name of the copyright holder nor the names of its * contributors may be used to endorse or promote products derived from this * software without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE * POSSIBILITY OF SUCH DAMAGE. */ /* * Copyright (c) 2016-2018, Klara Inc. * Copyright (c) 2016-2018, Allan Jude * Copyright (c) 2018-2020, Sebastian Gottschall * Copyright (c) 2019-2020, Michael Niewöhner * Copyright (c) 2020, The FreeBSD Foundation [1] * * [1] Portions of this software were developed by Allan Jude * under sponsorship from the FreeBSD Foundation. */ #ifndef _ZFS_ZSTD_H #define _ZFS_ZSTD_H #ifdef __cplusplus extern "C" { #endif /* * ZSTD block header * NOTE: all fields in this header are in big endian order. */ typedef struct zfs_zstd_header { /* Compressed size of data */ uint32_t c_len; /* * Version and compression level * We used to use a union to reference compression level * and version easily, but as it turns out, relying on the * ordering of bitfields is not remotely portable. * So now we have get/set functions in zfs_zstd.c for * manipulating this in just the right way forever. */ uint32_t raw_version_level; char data[]; } zfs_zstdhdr_t; /* * Simple struct to pass the data from raw_version_level around. */ typedef struct zfs_zstd_meta { uint8_t level; uint32_t version; } zfs_zstdmeta_t; /* * kstat helper macros */ #define ZSTDSTAT(stat) (zstd_stats.stat.value.ui64) #define ZSTDSTAT_ZERO(stat) \ atomic_store_64(&zstd_stats.stat.value.ui64, 0) #define ZSTDSTAT_ADD(stat, val) \ atomic_add_64(&zstd_stats.stat.value.ui64, (val)) #define ZSTDSTAT_SUB(stat, val) \ atomic_sub_64(&zstd_stats.stat.value.ui64, (val)) #define ZSTDSTAT_BUMP(stat) ZSTDSTAT_ADD(stat, 1) /* (de)init for user space / kernel emulation */ int zstd_init(void); void zstd_fini(void); -size_t zfs_zstd_compress(void *s_start, void *d_start, size_t s_len, +size_t zfs_zstd_compress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int level); int zfs_zstd_get_level(void *s_start, size_t s_len, uint8_t *level); -int zfs_zstd_decompress_level(void *s_start, void *d_start, size_t s_len, +int zfs_zstd_decompress_level(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, uint8_t *level); -int zfs_zstd_decompress(void *s_start, void *d_start, size_t s_len, +int zfs_zstd_decompress(abd_t *src, abd_t *dst, size_t s_len, size_t d_len, int n); void zfs_zstd_cache_reap_now(void); /* * So, the reason we have all these complicated set/get functions is that * originally, in the zstd "header" we wrote out to disk, we used a 32-bit * bitfield to store the "level" (8 bits) and "version" (24 bits). * * Unfortunately, bitfields make few promises about how they're arranged in * memory... * * By way of example, if we were using version 1.4.5 and level 3, it'd be * level = 0x03, version = 10405/0x0028A5, which gets broken into Vhigh = 0x00, * Vmid = 0x28, Vlow = 0xA5. We include these positions below to help follow * which data winds up where. * * As a consequence, we wound up with little endian platforms with a layout * like this in memory: * * 0 8 16 24 32 * +-------+-------+-------+-------+ * | Vlow | Vmid | Vhigh | level | * +-------+-------+-------+-------+ * =A5 =28 =00 =03 * * ...and then, after being run through BE_32(), serializing this out to * disk: * * 0 8 16 24 32 * +-------+-------+-------+-------+ * | level | Vhigh | Vmid | Vlow | * +-------+-------+-------+-------+ * =03 =00 =28 =A5 * * while on big-endian systems, since BE_32() is a noop there, both in * memory and on disk, we wind up with: * * 0 8 16 24 32 * +-------+-------+-------+-------+ * | Vhigh | Vmid | Vlow | level | * +-------+-------+-------+-------+ * =00 =28 =A5 =03 * * (Vhigh is always 0 until version exceeds 6.55.35. Vmid and Vlow are the * other two bytes of the "version" data.) * * So now we use the BF32_SET macros to get consistent behavior (the * ondisk LE encoding, since x86 currently rules the world) across * platforms, but the "get" behavior requires that we check each of the * bytes in the aforementioned former-bitfield for 0x00, and from there, * we can know which possible layout we're dealing with. (Only the two * that have been observed in the wild are illustrated above, but handlers * for all 4 positions of 0x00 are implemented. */ static inline void zfs_get_hdrmeta(const zfs_zstdhdr_t *blob, zfs_zstdmeta_t *res) { uint32_t raw = blob->raw_version_level; uint8_t findme = 0xff; int shift; for (shift = 0; shift < 4; shift++) { findme = BF32_GET(raw, 8*shift, 8); if (findme == 0) break; } switch (shift) { case 0: res->level = BF32_GET(raw, 24, 8); res->version = BSWAP_32(raw); res->version = BF32_GET(res->version, 8, 24); break; case 1: res->level = BF32_GET(raw, 0, 8); res->version = BSWAP_32(raw); res->version = BF32_GET(res->version, 0, 24); break; case 2: res->level = BF32_GET(raw, 24, 8); res->version = BF32_GET(raw, 0, 24); break; case 3: res->level = BF32_GET(raw, 0, 8); res->version = BF32_GET(raw, 8, 24); break; default: res->level = 0; res->version = 0; break; } } static inline uint8_t zfs_get_hdrlevel(const zfs_zstdhdr_t *blob) { uint8_t level = 0; zfs_zstdmeta_t res; zfs_get_hdrmeta(blob, &res); level = res.level; return (level); } static inline uint32_t zfs_get_hdrversion(const zfs_zstdhdr_t *blob) { uint32_t version = 0; zfs_zstdmeta_t res; zfs_get_hdrmeta(blob, &res); version = res.version; return (version); } static inline void zfs_set_hdrversion(zfs_zstdhdr_t *blob, uint32_t version) { /* cppcheck-suppress syntaxError */ BF32_SET(blob->raw_version_level, 0, 24, version); } static inline void zfs_set_hdrlevel(zfs_zstdhdr_t *blob, uint8_t level) { /* cppcheck-suppress syntaxError */ BF32_SET(blob->raw_version_level, 24, 8, level); } #ifdef __cplusplus } #endif #endif /* _ZFS_ZSTD_H */ diff --git a/module/zfs/ddt_zap.c b/module/zfs/ddt_zap.c index 8e78ec3277c7..e96984b86f0c 100644 --- a/module/zfs/ddt_zap.c +++ b/module/zfs/ddt_zap.c @@ -1,258 +1,262 @@ /* * 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) 2009, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright (c) 2018 by Delphix. All rights reserved. * Copyright (c) 2023, Klara Inc. */ #include #include #include #include #include #include #include #include static unsigned int ddt_zap_default_bs = 15; static unsigned int ddt_zap_default_ibs = 15; #define DDT_ZAP_COMPRESS_BYTEORDER_MASK 0x80 #define DDT_ZAP_COMPRESS_FUNCTION_MASK 0x7f #define DDT_KEY_WORDS (sizeof (ddt_key_t) / sizeof (uint64_t)) static size_t ddt_zap_compress(const void *src, uchar_t *dst, size_t s_len, size_t d_len) { uchar_t *version = dst++; int cpfunc = ZIO_COMPRESS_ZLE; zio_compress_info_t *ci = &zio_compress_table[cpfunc]; size_t c_len; ASSERT3U(d_len, >=, s_len + 1); /* no compression plus version byte */ /* Call compress function directly to avoid hole detection. */ - c_len = ci->ci_compress((void *)src, dst, s_len, d_len - 1, - ci->ci_level); + abd_t sabd, dabd; + abd_get_from_buf_struct(&sabd, (void *)src, s_len); + abd_get_from_buf_struct(&dabd, dst, d_len); + c_len = ci->ci_compress(&sabd, &dabd, s_len, d_len - 1, ci->ci_level); + abd_free(&dabd); + abd_free(&sabd); if (c_len == s_len) { cpfunc = ZIO_COMPRESS_OFF; memcpy(dst, src, s_len); } *version = cpfunc; if (ZFS_HOST_BYTEORDER) *version |= DDT_ZAP_COMPRESS_BYTEORDER_MASK; return (c_len + 1); } static void ddt_zap_decompress(uchar_t *src, void *dst, size_t s_len, size_t d_len) { uchar_t version = *src++; int cpfunc = version & DDT_ZAP_COMPRESS_FUNCTION_MASK; if (zio_compress_table[cpfunc].ci_decompress == NULL) { memcpy(dst, src, d_len); return; } abd_t sabd; abd_get_from_buf_struct(&sabd, src, s_len); VERIFY0(zio_decompress_data(cpfunc, &sabd, dst, s_len, d_len, NULL)); abd_free(&sabd); if (((version & DDT_ZAP_COMPRESS_BYTEORDER_MASK) != 0) != (ZFS_HOST_BYTEORDER != 0)) byteswap_uint64_array(dst, d_len); } static int ddt_zap_create(objset_t *os, uint64_t *objectp, dmu_tx_t *tx, boolean_t prehash) { zap_flags_t flags = ZAP_FLAG_HASH64 | ZAP_FLAG_UINT64_KEY; if (prehash) flags |= ZAP_FLAG_PRE_HASHED_KEY; *objectp = zap_create_flags(os, 0, flags, DMU_OT_DDT_ZAP, ddt_zap_default_bs, ddt_zap_default_ibs, DMU_OT_NONE, 0, tx); if (*objectp == 0) return (SET_ERROR(ENOTSUP)); return (0); } static int ddt_zap_destroy(objset_t *os, uint64_t object, dmu_tx_t *tx) { return (zap_destroy(os, object, tx)); } static int ddt_zap_lookup(objset_t *os, uint64_t object, const ddt_key_t *ddk, void *phys, size_t psize) { uchar_t *cbuf; uint64_t one, csize; int error; error = zap_length_uint64(os, object, (uint64_t *)ddk, DDT_KEY_WORDS, &one, &csize); if (error) return (error); ASSERT3U(one, ==, 1); ASSERT3U(csize, <=, psize + 1); cbuf = kmem_alloc(csize, KM_SLEEP); error = zap_lookup_uint64(os, object, (uint64_t *)ddk, DDT_KEY_WORDS, 1, csize, cbuf); if (error == 0) ddt_zap_decompress(cbuf, phys, csize, psize); kmem_free(cbuf, csize); return (error); } static int ddt_zap_contains(objset_t *os, uint64_t object, const ddt_key_t *ddk) { return (zap_length_uint64(os, object, (uint64_t *)ddk, DDT_KEY_WORDS, NULL, NULL)); } static void ddt_zap_prefetch(objset_t *os, uint64_t object, const ddt_key_t *ddk) { (void) zap_prefetch_uint64(os, object, (uint64_t *)ddk, DDT_KEY_WORDS); } static void ddt_zap_prefetch_all(objset_t *os, uint64_t object) { (void) zap_prefetch_object(os, object); } static int ddt_zap_update(objset_t *os, uint64_t object, const ddt_key_t *ddk, const void *phys, size_t psize, dmu_tx_t *tx) { const size_t cbuf_size = psize + 1; uchar_t *cbuf = kmem_alloc(cbuf_size, KM_SLEEP); uint64_t csize = ddt_zap_compress(phys, cbuf, psize, cbuf_size); int error = zap_update_uint64(os, object, (uint64_t *)ddk, DDT_KEY_WORDS, 1, csize, cbuf, tx); kmem_free(cbuf, cbuf_size); return (error); } static int ddt_zap_remove(objset_t *os, uint64_t object, const ddt_key_t *ddk, dmu_tx_t *tx) { return (zap_remove_uint64(os, object, (uint64_t *)ddk, DDT_KEY_WORDS, tx)); } static int ddt_zap_walk(objset_t *os, uint64_t object, uint64_t *walk, ddt_key_t *ddk, void *phys, size_t psize) { zap_cursor_t zc; zap_attribute_t za; int error; if (*walk == 0) { /* * We don't want to prefetch the entire ZAP object, because * it can be enormous. Also the primary use of DDT iteration * is for scrubbing, in which case we will be issuing many * scrub I/Os for each ZAP block that we read in, so * reading the ZAP is unlikely to be the bottleneck. */ zap_cursor_init_noprefetch(&zc, os, object); } else { zap_cursor_init_serialized(&zc, os, object, *walk); } if ((error = zap_cursor_retrieve(&zc, &za)) == 0) { uint64_t csize = za.za_num_integers; ASSERT3U(za.za_integer_length, ==, 1); ASSERT3U(csize, <=, psize + 1); uchar_t *cbuf = kmem_alloc(csize, KM_SLEEP); error = zap_lookup_uint64(os, object, (uint64_t *)za.za_name, DDT_KEY_WORDS, 1, csize, cbuf); ASSERT0(error); if (error == 0) { ddt_zap_decompress(cbuf, phys, csize, psize); *ddk = *(ddt_key_t *)za.za_name; } kmem_free(cbuf, csize); zap_cursor_advance(&zc); *walk = zap_cursor_serialize(&zc); } zap_cursor_fini(&zc); return (error); } static int ddt_zap_count(objset_t *os, uint64_t object, uint64_t *count) { return (zap_count(os, object, count)); } const ddt_ops_t ddt_zap_ops = { "zap", ddt_zap_create, ddt_zap_destroy, ddt_zap_lookup, ddt_zap_contains, ddt_zap_prefetch, ddt_zap_prefetch_all, ddt_zap_update, ddt_zap_remove, ddt_zap_walk, ddt_zap_count, }; /* BEGIN CSTYLED */ ZFS_MODULE_PARAM(zfs_dedup, , ddt_zap_default_bs, UINT, ZMOD_RW, "DDT ZAP leaf blockshift"); ZFS_MODULE_PARAM(zfs_dedup, , ddt_zap_default_ibs, UINT, ZMOD_RW, "DDT ZAP indirect blockshift"); /* END CSTYLED */ diff --git a/module/zfs/dsl_dataset.c b/module/zfs/dsl_dataset.c index e62ecdb259f7..042725b235d0 100644 --- a/module/zfs/dsl_dataset.c +++ b/module/zfs/dsl_dataset.c @@ -1,5020 +1,5025 @@ /* * 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, 2020 by Delphix. All rights reserved. * Copyright (c) 2014, Joyent, Inc. All rights reserved. * Copyright (c) 2014 RackTop Systems. * Copyright (c) 2014 Spectra Logic Corporation, All rights reserved. * Copyright (c) 2016 Actifio, Inc. All rights reserved. * Copyright 2016, OmniTI Computer Consulting, Inc. All rights reserved. * Copyright 2017 Nexenta Systems, Inc. * Copyright (c) 2019, Klara Inc. * Copyright (c) 2019, Allan Jude * Copyright (c) 2020 The FreeBSD Foundation [1] * * [1] Portions of this software were developed by Allan Jude * under sponsorship from the FreeBSD Foundation. */ #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 /* * The SPA supports block sizes up to 16MB. However, very large blocks * can have an impact on i/o latency (e.g. tying up a spinning disk for * ~300ms), and also potentially on the memory allocator. Therefore, * we did not allow the recordsize to be set larger than zfs_max_recordsize * (former default: 1MB). Larger blocks could be created by changing this * tunable, and pools with larger blocks could always be imported and used, * regardless of this setting. * * We do, however, still limit it by default to 1M on x86_32, because Linux's * 3/1 memory split doesn't leave much room for 16M chunks. */ #ifdef _ILP32 uint_t zfs_max_recordsize = 1 * 1024 * 1024; #else uint_t zfs_max_recordsize = 16 * 1024 * 1024; #endif static int zfs_allow_redacted_dataset_mount = 0; int zfs_snapshot_history_enabled = 1; #define SWITCH64(x, y) \ { \ uint64_t __tmp = (x); \ (x) = (y); \ (y) = __tmp; \ } #define DS_REF_MAX (1ULL << 62) static void dsl_dataset_set_remap_deadlist_object(dsl_dataset_t *ds, uint64_t obj, dmu_tx_t *tx); static void dsl_dataset_unset_remap_deadlist_object(dsl_dataset_t *ds, dmu_tx_t *tx); static void unload_zfeature(dsl_dataset_t *ds, spa_feature_t f); extern uint_t spa_asize_inflation; static zil_header_t zero_zil; /* * Figure out how much of this delta should be propagated to the dsl_dir * layer. If there's a refreservation, that space has already been * partially accounted for in our ancestors. */ static int64_t parent_delta(dsl_dataset_t *ds, int64_t delta) { dsl_dataset_phys_t *ds_phys; uint64_t old_bytes, new_bytes; if (ds->ds_reserved == 0) return (delta); ds_phys = dsl_dataset_phys(ds); old_bytes = MAX(ds_phys->ds_unique_bytes, ds->ds_reserved); new_bytes = MAX(ds_phys->ds_unique_bytes + delta, ds->ds_reserved); ASSERT3U(ABS((int64_t)(new_bytes - old_bytes)), <=, ABS(delta)); return (new_bytes - old_bytes); } void dsl_dataset_block_born(dsl_dataset_t *ds, const blkptr_t *bp, dmu_tx_t *tx) { spa_t *spa = dmu_tx_pool(tx)->dp_spa; int used = bp_get_dsize_sync(spa, bp); int compressed = BP_GET_PSIZE(bp); int uncompressed = BP_GET_UCSIZE(bp); int64_t delta; spa_feature_t f; dprintf_bp(bp, "ds=%p", ds); ASSERT(dmu_tx_is_syncing(tx)); /* It could have been compressed away to nothing */ if (BP_IS_HOLE(bp) || BP_IS_REDACTED(bp)) return; ASSERT(BP_GET_TYPE(bp) != DMU_OT_NONE); ASSERT(DMU_OT_IS_VALID(BP_GET_TYPE(bp))); if (ds == NULL) { dsl_pool_mos_diduse_space(tx->tx_pool, used, compressed, uncompressed); return; } ASSERT3U(BP_GET_LOGICAL_BIRTH(bp), >, dsl_dataset_phys(ds)->ds_prev_snap_txg); dmu_buf_will_dirty(ds->ds_dbuf, tx); mutex_enter(&ds->ds_lock); delta = parent_delta(ds, used); dsl_dataset_phys(ds)->ds_referenced_bytes += used; dsl_dataset_phys(ds)->ds_compressed_bytes += compressed; dsl_dataset_phys(ds)->ds_uncompressed_bytes += uncompressed; dsl_dataset_phys(ds)->ds_unique_bytes += used; if (BP_GET_LSIZE(bp) > SPA_OLD_MAXBLOCKSIZE) { ds->ds_feature_activation[SPA_FEATURE_LARGE_BLOCKS] = (void *)B_TRUE; } f = zio_checksum_to_feature(BP_GET_CHECKSUM(bp)); if (f != SPA_FEATURE_NONE) { ASSERT3S(spa_feature_table[f].fi_type, ==, ZFEATURE_TYPE_BOOLEAN); ds->ds_feature_activation[f] = (void *)B_TRUE; } f = zio_compress_to_feature(BP_GET_COMPRESS(bp)); if (f != SPA_FEATURE_NONE) { ASSERT3S(spa_feature_table[f].fi_type, ==, ZFEATURE_TYPE_BOOLEAN); ds->ds_feature_activation[f] = (void *)B_TRUE; } /* * Track block for livelist, but ignore embedded blocks because * they do not need to be freed. */ if (dsl_deadlist_is_open(&ds->ds_dir->dd_livelist) && BP_GET_LOGICAL_BIRTH(bp) > ds->ds_dir->dd_origin_txg && !(BP_IS_EMBEDDED(bp))) { ASSERT(dsl_dir_is_clone(ds->ds_dir)); ASSERT(spa_feature_is_enabled(spa, SPA_FEATURE_LIVELIST)); bplist_append(&ds->ds_dir->dd_pending_allocs, bp); } mutex_exit(&ds->ds_lock); dsl_dir_diduse_transfer_space(ds->ds_dir, delta, compressed, uncompressed, used, DD_USED_REFRSRV, DD_USED_HEAD, tx); } /* * Called when the specified segment has been remapped, and is thus no * longer referenced in the head dataset. The vdev must be indirect. * * If the segment is referenced by a snapshot, put it on the remap deadlist. * Otherwise, add this segment to the obsolete spacemap. */ void dsl_dataset_block_remapped(dsl_dataset_t *ds, uint64_t vdev, uint64_t offset, uint64_t size, uint64_t birth, dmu_tx_t *tx) { spa_t *spa = ds->ds_dir->dd_pool->dp_spa; ASSERT(dmu_tx_is_syncing(tx)); ASSERT(birth <= tx->tx_txg); ASSERT(!ds->ds_is_snapshot); if (birth > dsl_dataset_phys(ds)->ds_prev_snap_txg) { spa_vdev_indirect_mark_obsolete(spa, vdev, offset, size, tx); } else { blkptr_t fakebp; dva_t *dva = &fakebp.blk_dva[0]; ASSERT(ds != NULL); mutex_enter(&ds->ds_remap_deadlist_lock); if (!dsl_dataset_remap_deadlist_exists(ds)) { dsl_dataset_create_remap_deadlist(ds, tx); } mutex_exit(&ds->ds_remap_deadlist_lock); BP_ZERO(&fakebp); BP_SET_LOGICAL_BIRTH(&fakebp, birth); DVA_SET_VDEV(dva, vdev); DVA_SET_OFFSET(dva, offset); DVA_SET_ASIZE(dva, size); dsl_deadlist_insert(&ds->ds_remap_deadlist, &fakebp, B_FALSE, tx); } } int dsl_dataset_block_kill(dsl_dataset_t *ds, const blkptr_t *bp, dmu_tx_t *tx, boolean_t async) { spa_t *spa = dmu_tx_pool(tx)->dp_spa; int used = bp_get_dsize_sync(spa, bp); int compressed = BP_GET_PSIZE(bp); int uncompressed = BP_GET_UCSIZE(bp); if (BP_IS_HOLE(bp) || BP_IS_REDACTED(bp)) return (0); ASSERT(dmu_tx_is_syncing(tx)); ASSERT(BP_GET_LOGICAL_BIRTH(bp) <= tx->tx_txg); if (ds == NULL) { dsl_free(tx->tx_pool, tx->tx_txg, bp); dsl_pool_mos_diduse_space(tx->tx_pool, -used, -compressed, -uncompressed); return (used); } ASSERT3P(tx->tx_pool, ==, ds->ds_dir->dd_pool); ASSERT(!ds->ds_is_snapshot); dmu_buf_will_dirty(ds->ds_dbuf, tx); /* * Track block for livelist, but ignore embedded blocks because * they do not need to be freed. */ if (dsl_deadlist_is_open(&ds->ds_dir->dd_livelist) && BP_GET_LOGICAL_BIRTH(bp) > ds->ds_dir->dd_origin_txg && !(BP_IS_EMBEDDED(bp))) { ASSERT(dsl_dir_is_clone(ds->ds_dir)); ASSERT(spa_feature_is_enabled(spa, SPA_FEATURE_LIVELIST)); bplist_append(&ds->ds_dir->dd_pending_frees, bp); } if (BP_GET_LOGICAL_BIRTH(bp) > dsl_dataset_phys(ds)->ds_prev_snap_txg) { int64_t delta; dprintf_bp(bp, "freeing ds=%llu", (u_longlong_t)ds->ds_object); dsl_free(tx->tx_pool, tx->tx_txg, bp); mutex_enter(&ds->ds_lock); ASSERT(dsl_dataset_phys(ds)->ds_unique_bytes >= used || !DS_UNIQUE_IS_ACCURATE(ds)); delta = parent_delta(ds, -used); dsl_dataset_phys(ds)->ds_unique_bytes -= used; mutex_exit(&ds->ds_lock); dsl_dir_diduse_transfer_space(ds->ds_dir, delta, -compressed, -uncompressed, -used, DD_USED_REFRSRV, DD_USED_HEAD, tx); } else { dprintf_bp(bp, "putting on dead list: %s", ""); if (async) { /* * We are here as part of zio's write done callback, * which means we're a zio interrupt thread. We can't * call dsl_deadlist_insert() now because it may block * waiting for I/O. Instead, put bp on the deferred * queue and let dsl_pool_sync() finish the job. */ bplist_append(&ds->ds_pending_deadlist, bp); } else { dsl_deadlist_insert(&ds->ds_deadlist, bp, B_FALSE, tx); } ASSERT3U(ds->ds_prev->ds_object, ==, dsl_dataset_phys(ds)->ds_prev_snap_obj); ASSERT(dsl_dataset_phys(ds->ds_prev)->ds_num_children > 0); /* if (logical birth > prev prev snap txg) prev unique += bs */ if (dsl_dataset_phys(ds->ds_prev)->ds_next_snap_obj == ds->ds_object && BP_GET_LOGICAL_BIRTH(bp) > dsl_dataset_phys(ds->ds_prev)->ds_prev_snap_txg) { dmu_buf_will_dirty(ds->ds_prev->ds_dbuf, tx); mutex_enter(&ds->ds_prev->ds_lock); dsl_dataset_phys(ds->ds_prev)->ds_unique_bytes += used; mutex_exit(&ds->ds_prev->ds_lock); } if (BP_GET_LOGICAL_BIRTH(bp) > ds->ds_dir->dd_origin_txg) { dsl_dir_transfer_space(ds->ds_dir, used, DD_USED_HEAD, DD_USED_SNAP, tx); } } dsl_bookmark_block_killed(ds, bp, tx); mutex_enter(&ds->ds_lock); ASSERT3U(dsl_dataset_phys(ds)->ds_referenced_bytes, >=, used); dsl_dataset_phys(ds)->ds_referenced_bytes -= used; ASSERT3U(dsl_dataset_phys(ds)->ds_compressed_bytes, >=, compressed); dsl_dataset_phys(ds)->ds_compressed_bytes -= compressed; ASSERT3U(dsl_dataset_phys(ds)->ds_uncompressed_bytes, >=, uncompressed); dsl_dataset_phys(ds)->ds_uncompressed_bytes -= uncompressed; mutex_exit(&ds->ds_lock); return (used); } struct feature_type_uint64_array_arg { uint64_t length; uint64_t *array; }; static void unload_zfeature(dsl_dataset_t *ds, spa_feature_t f) { switch (spa_feature_table[f].fi_type) { case ZFEATURE_TYPE_BOOLEAN: break; case ZFEATURE_TYPE_UINT64_ARRAY: { struct feature_type_uint64_array_arg *ftuaa = ds->ds_feature[f]; kmem_free(ftuaa->array, ftuaa->length * sizeof (uint64_t)); kmem_free(ftuaa, sizeof (*ftuaa)); break; } default: panic("Invalid zfeature type %d", spa_feature_table[f].fi_type); } } static int load_zfeature(objset_t *mos, dsl_dataset_t *ds, spa_feature_t f) { int err = 0; switch (spa_feature_table[f].fi_type) { case ZFEATURE_TYPE_BOOLEAN: err = zap_contains(mos, ds->ds_object, spa_feature_table[f].fi_guid); if (err == 0) { ds->ds_feature[f] = (void *)B_TRUE; } else { ASSERT3U(err, ==, ENOENT); err = 0; } break; case ZFEATURE_TYPE_UINT64_ARRAY: { uint64_t int_size, num_int; uint64_t *data; err = zap_length(mos, ds->ds_object, spa_feature_table[f].fi_guid, &int_size, &num_int); if (err != 0) { ASSERT3U(err, ==, ENOENT); err = 0; break; } ASSERT3U(int_size, ==, sizeof (uint64_t)); data = kmem_alloc(int_size * num_int, KM_SLEEP); VERIFY0(zap_lookup(mos, ds->ds_object, spa_feature_table[f].fi_guid, int_size, num_int, data)); struct feature_type_uint64_array_arg *ftuaa = kmem_alloc(sizeof (*ftuaa), KM_SLEEP); ftuaa->length = num_int; ftuaa->array = data; ds->ds_feature[f] = ftuaa; break; } default: panic("Invalid zfeature type %d", spa_feature_table[f].fi_type); } return (err); } /* * We have to release the fsid synchronously or we risk that a subsequent * mount of the same dataset will fail to unique_insert the fsid. This * failure would manifest itself as the fsid of this dataset changing * between mounts which makes NFS clients quite unhappy. */ static void dsl_dataset_evict_sync(void *dbu) { dsl_dataset_t *ds = dbu; ASSERT(ds->ds_owner == NULL); unique_remove(ds->ds_fsid_guid); } static void dsl_dataset_evict_async(void *dbu) { dsl_dataset_t *ds = dbu; ASSERT(ds->ds_owner == NULL); ds->ds_dbuf = NULL; if (ds->ds_objset != NULL) dmu_objset_evict(ds->ds_objset); if (ds->ds_prev) { dsl_dataset_rele(ds->ds_prev, ds); ds->ds_prev = NULL; } dsl_bookmark_fini_ds(ds); bplist_destroy(&ds->ds_pending_deadlist); if (dsl_deadlist_is_open(&ds->ds_deadlist)) dsl_deadlist_close(&ds->ds_deadlist); if (dsl_deadlist_is_open(&ds->ds_remap_deadlist)) dsl_deadlist_close(&ds->ds_remap_deadlist); if (ds->ds_dir) dsl_dir_async_rele(ds->ds_dir, ds); ASSERT(!list_link_active(&ds->ds_synced_link)); for (spa_feature_t f = 0; f < SPA_FEATURES; f++) { if (dsl_dataset_feature_is_active(ds, f)) unload_zfeature(ds, f); } list_destroy(&ds->ds_prop_cbs); mutex_destroy(&ds->ds_lock); mutex_destroy(&ds->ds_opening_lock); mutex_destroy(&ds->ds_sendstream_lock); mutex_destroy(&ds->ds_remap_deadlist_lock); zfs_refcount_destroy(&ds->ds_longholds); rrw_destroy(&ds->ds_bp_rwlock); kmem_free(ds, sizeof (dsl_dataset_t)); } int dsl_dataset_get_snapname(dsl_dataset_t *ds) { dsl_dataset_phys_t *headphys; int err; dmu_buf_t *headdbuf; dsl_pool_t *dp = ds->ds_dir->dd_pool; objset_t *mos = dp->dp_meta_objset; if (ds->ds_snapname[0]) return (0); if (dsl_dataset_phys(ds)->ds_next_snap_obj == 0) return (0); err = dmu_bonus_hold(mos, dsl_dir_phys(ds->ds_dir)->dd_head_dataset_obj, FTAG, &headdbuf); if (err != 0) return (err); headphys = headdbuf->db_data; err = zap_value_search(dp->dp_meta_objset, headphys->ds_snapnames_zapobj, ds->ds_object, 0, ds->ds_snapname); if (err != 0 && zfs_recover == B_TRUE) { err = 0; (void) snprintf(ds->ds_snapname, sizeof (ds->ds_snapname), "SNAPOBJ=%llu-ERR=%d", (unsigned long long)ds->ds_object, err); } dmu_buf_rele(headdbuf, FTAG); return (err); } int dsl_dataset_snap_lookup(dsl_dataset_t *ds, const char *name, uint64_t *value) { objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset; uint64_t snapobj = dsl_dataset_phys(ds)->ds_snapnames_zapobj; matchtype_t mt = 0; int err; if (dsl_dataset_phys(ds)->ds_flags & DS_FLAG_CI_DATASET) mt = MT_NORMALIZE; err = zap_lookup_norm(mos, snapobj, name, 8, 1, value, mt, NULL, 0, NULL); if (err == ENOTSUP && (mt & MT_NORMALIZE)) err = zap_lookup(mos, snapobj, name, 8, 1, value); return (err); } int dsl_dataset_snap_remove(dsl_dataset_t *ds, const char *name, dmu_tx_t *tx, boolean_t adj_cnt) { objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset; uint64_t snapobj = dsl_dataset_phys(ds)->ds_snapnames_zapobj; matchtype_t mt = 0; int err; dsl_dir_snap_cmtime_update(ds->ds_dir, tx); if (dsl_dataset_phys(ds)->ds_flags & DS_FLAG_CI_DATASET) mt = MT_NORMALIZE; err = zap_remove_norm(mos, snapobj, name, mt, tx); if (err == ENOTSUP && (mt & MT_NORMALIZE)) err = zap_remove(mos, snapobj, name, tx); if (err == 0 && adj_cnt) dsl_fs_ss_count_adjust(ds->ds_dir, -1, DD_FIELD_SNAPSHOT_COUNT, tx); return (err); } boolean_t dsl_dataset_try_add_ref(dsl_pool_t *dp, dsl_dataset_t *ds, const void *tag) { dmu_buf_t *dbuf = ds->ds_dbuf; boolean_t result = B_FALSE; if (dbuf != NULL && dmu_buf_try_add_ref(dbuf, dp->dp_meta_objset, ds->ds_object, DMU_BONUS_BLKID, tag)) { if (ds == dmu_buf_get_user(dbuf)) result = B_TRUE; else dmu_buf_rele(dbuf, tag); } return (result); } int dsl_dataset_hold_obj(dsl_pool_t *dp, uint64_t dsobj, const void *tag, dsl_dataset_t **dsp) { objset_t *mos = dp->dp_meta_objset; dmu_buf_t *dbuf; dsl_dataset_t *ds; int err; dmu_object_info_t doi; ASSERT(dsl_pool_config_held(dp)); err = dmu_bonus_hold(mos, dsobj, tag, &dbuf); if (err != 0) return (err); /* Make sure dsobj has the correct object type. */ dmu_object_info_from_db(dbuf, &doi); if (doi.doi_bonus_type != DMU_OT_DSL_DATASET) { dmu_buf_rele(dbuf, tag); return (SET_ERROR(EINVAL)); } ds = dmu_buf_get_user(dbuf); if (ds == NULL) { dsl_dataset_t *winner = NULL; ds = kmem_zalloc(sizeof (dsl_dataset_t), KM_SLEEP); ds->ds_dbuf = dbuf; ds->ds_object = dsobj; ds->ds_is_snapshot = dsl_dataset_phys(ds)->ds_num_children != 0; list_link_init(&ds->ds_synced_link); err = dsl_dir_hold_obj(dp, dsl_dataset_phys(ds)->ds_dir_obj, NULL, ds, &ds->ds_dir); if (err != 0) { kmem_free(ds, sizeof (dsl_dataset_t)); dmu_buf_rele(dbuf, tag); return (err); } mutex_init(&ds->ds_lock, NULL, MUTEX_DEFAULT, NULL); mutex_init(&ds->ds_opening_lock, NULL, MUTEX_DEFAULT, NULL); mutex_init(&ds->ds_sendstream_lock, NULL, MUTEX_DEFAULT, NULL); mutex_init(&ds->ds_remap_deadlist_lock, NULL, MUTEX_DEFAULT, NULL); rrw_init(&ds->ds_bp_rwlock, B_FALSE); zfs_refcount_create(&ds->ds_longholds); bplist_create(&ds->ds_pending_deadlist); list_create(&ds->ds_sendstreams, sizeof (dmu_sendstatus_t), offsetof(dmu_sendstatus_t, dss_link)); list_create(&ds->ds_prop_cbs, sizeof (dsl_prop_cb_record_t), offsetof(dsl_prop_cb_record_t, cbr_ds_node)); if (doi.doi_type == DMU_OTN_ZAP_METADATA) { spa_feature_t f; for (f = 0; f < SPA_FEATURES; f++) { if (!(spa_feature_table[f].fi_flags & ZFEATURE_FLAG_PER_DATASET)) continue; err = load_zfeature(mos, ds, f); } } if (!ds->ds_is_snapshot) { ds->ds_snapname[0] = '\0'; if (dsl_dataset_phys(ds)->ds_prev_snap_obj != 0) { err = dsl_dataset_hold_obj(dp, dsl_dataset_phys(ds)->ds_prev_snap_obj, ds, &ds->ds_prev); } if (err != 0) goto after_dsl_bookmark_fini; err = dsl_bookmark_init_ds(ds); } else { if (zfs_flags & ZFS_DEBUG_SNAPNAMES) err = dsl_dataset_get_snapname(ds); if (err == 0 && dsl_dataset_phys(ds)->ds_userrefs_obj != 0) { err = zap_count( ds->ds_dir->dd_pool->dp_meta_objset, dsl_dataset_phys(ds)->ds_userrefs_obj, &ds->ds_userrefs); } } if (err == 0 && !ds->ds_is_snapshot) { err = dsl_prop_get_int_ds(ds, zfs_prop_to_name(ZFS_PROP_REFRESERVATION), &ds->ds_reserved); if (err == 0) { err = dsl_prop_get_int_ds(ds, zfs_prop_to_name(ZFS_PROP_REFQUOTA), &ds->ds_quota); } } else { ds->ds_reserved = ds->ds_quota = 0; } if (err == 0 && ds->ds_dir->dd_crypto_obj != 0 && ds->ds_is_snapshot && zap_contains(mos, dsobj, DS_FIELD_IVSET_GUID) != 0) { dp->dp_spa->spa_errata = ZPOOL_ERRATA_ZOL_8308_ENCRYPTION; } dsl_deadlist_open(&ds->ds_deadlist, mos, dsl_dataset_phys(ds)->ds_deadlist_obj); uint64_t remap_deadlist_obj = dsl_dataset_get_remap_deadlist_object(ds); if (remap_deadlist_obj != 0) { dsl_deadlist_open(&ds->ds_remap_deadlist, mos, remap_deadlist_obj); } dmu_buf_init_user(&ds->ds_dbu, dsl_dataset_evict_sync, dsl_dataset_evict_async, &ds->ds_dbuf); if (err == 0) winner = dmu_buf_set_user_ie(dbuf, &ds->ds_dbu); if (err != 0 || winner != NULL) { dsl_deadlist_close(&ds->ds_deadlist); if (dsl_deadlist_is_open(&ds->ds_remap_deadlist)) dsl_deadlist_close(&ds->ds_remap_deadlist); dsl_bookmark_fini_ds(ds); after_dsl_bookmark_fini: if (ds->ds_prev) dsl_dataset_rele(ds->ds_prev, ds); dsl_dir_rele(ds->ds_dir, ds); for (spa_feature_t f = 0; f < SPA_FEATURES; f++) { if (dsl_dataset_feature_is_active(ds, f)) unload_zfeature(ds, f); } list_destroy(&ds->ds_prop_cbs); list_destroy(&ds->ds_sendstreams); bplist_destroy(&ds->ds_pending_deadlist); mutex_destroy(&ds->ds_lock); mutex_destroy(&ds->ds_opening_lock); mutex_destroy(&ds->ds_sendstream_lock); mutex_destroy(&ds->ds_remap_deadlist_lock); zfs_refcount_destroy(&ds->ds_longholds); rrw_destroy(&ds->ds_bp_rwlock); kmem_free(ds, sizeof (dsl_dataset_t)); if (err != 0) { dmu_buf_rele(dbuf, tag); return (err); } ds = winner; } else { ds->ds_fsid_guid = unique_insert(dsl_dataset_phys(ds)->ds_fsid_guid); if (ds->ds_fsid_guid != dsl_dataset_phys(ds)->ds_fsid_guid) { zfs_dbgmsg("ds_fsid_guid changed from " "%llx to %llx for pool %s dataset id %llu", (long long) dsl_dataset_phys(ds)->ds_fsid_guid, (long long)ds->ds_fsid_guid, spa_name(dp->dp_spa), (u_longlong_t)dsobj); } } } ASSERT3P(ds->ds_dbuf, ==, dbuf); ASSERT3P(dsl_dataset_phys(ds), ==, dbuf->db_data); ASSERT(dsl_dataset_phys(ds)->ds_prev_snap_obj != 0 || spa_version(dp->dp_spa) < SPA_VERSION_ORIGIN || dp->dp_origin_snap == NULL || ds == dp->dp_origin_snap); *dsp = ds; return (0); } int dsl_dataset_create_key_mapping(dsl_dataset_t *ds) { dsl_dir_t *dd = ds->ds_dir; if (dd->dd_crypto_obj == 0) return (0); return (spa_keystore_create_mapping(dd->dd_pool->dp_spa, ds, ds, &ds->ds_key_mapping)); } int dsl_dataset_hold_obj_flags(dsl_pool_t *dp, uint64_t dsobj, ds_hold_flags_t flags, const void *tag, dsl_dataset_t **dsp) { int err; err = dsl_dataset_hold_obj(dp, dsobj, tag, dsp); if (err != 0) return (err); ASSERT3P(*dsp, !=, NULL); if (flags & DS_HOLD_FLAG_DECRYPT) { err = dsl_dataset_create_key_mapping(*dsp); if (err != 0) dsl_dataset_rele(*dsp, tag); } return (err); } int dsl_dataset_hold_flags(dsl_pool_t *dp, const char *name, ds_hold_flags_t flags, const void *tag, dsl_dataset_t **dsp) { dsl_dir_t *dd; const char *snapname; uint64_t obj; int err = 0; dsl_dataset_t *ds; err = dsl_dir_hold(dp, name, FTAG, &dd, &snapname); if (err != 0) return (err); ASSERT(dsl_pool_config_held(dp)); obj = dsl_dir_phys(dd)->dd_head_dataset_obj; if (obj != 0) err = dsl_dataset_hold_obj_flags(dp, obj, flags, tag, &ds); else err = SET_ERROR(ENOENT); /* we may be looking for a snapshot */ if (err == 0 && snapname != NULL) { dsl_dataset_t *snap_ds; if (*snapname++ != '@') { dsl_dataset_rele_flags(ds, flags, tag); dsl_dir_rele(dd, FTAG); return (SET_ERROR(ENOENT)); } dprintf("looking for snapshot '%s'\n", snapname); err = dsl_dataset_snap_lookup(ds, snapname, &obj); if (err == 0) { err = dsl_dataset_hold_obj_flags(dp, obj, flags, tag, &snap_ds); } dsl_dataset_rele_flags(ds, flags, tag); if (err == 0) { mutex_enter(&snap_ds->ds_lock); if (snap_ds->ds_snapname[0] == 0) (void) strlcpy(snap_ds->ds_snapname, snapname, sizeof (snap_ds->ds_snapname)); mutex_exit(&snap_ds->ds_lock); ds = snap_ds; } } if (err == 0) *dsp = ds; dsl_dir_rele(dd, FTAG); return (err); } int dsl_dataset_hold(dsl_pool_t *dp, const char *name, const void *tag, dsl_dataset_t **dsp) { return (dsl_dataset_hold_flags(dp, name, 0, tag, dsp)); } static int dsl_dataset_own_obj_impl(dsl_pool_t *dp, uint64_t dsobj, ds_hold_flags_t flags, const void *tag, boolean_t override, dsl_dataset_t **dsp) { int err = dsl_dataset_hold_obj_flags(dp, dsobj, flags, tag, dsp); if (err != 0) return (err); if (!dsl_dataset_tryown(*dsp, tag, override)) { dsl_dataset_rele_flags(*dsp, flags, tag); *dsp = NULL; return (SET_ERROR(EBUSY)); } return (0); } int dsl_dataset_own_obj(dsl_pool_t *dp, uint64_t dsobj, ds_hold_flags_t flags, const void *tag, dsl_dataset_t **dsp) { return (dsl_dataset_own_obj_impl(dp, dsobj, flags, tag, B_FALSE, dsp)); } int dsl_dataset_own_obj_force(dsl_pool_t *dp, uint64_t dsobj, ds_hold_flags_t flags, const void *tag, dsl_dataset_t **dsp) { return (dsl_dataset_own_obj_impl(dp, dsobj, flags, tag, B_TRUE, dsp)); } static int dsl_dataset_own_impl(dsl_pool_t *dp, const char *name, ds_hold_flags_t flags, const void *tag, boolean_t override, dsl_dataset_t **dsp) { int err = dsl_dataset_hold_flags(dp, name, flags, tag, dsp); if (err != 0) return (err); if (!dsl_dataset_tryown(*dsp, tag, override)) { dsl_dataset_rele_flags(*dsp, flags, tag); return (SET_ERROR(EBUSY)); } return (0); } int dsl_dataset_own_force(dsl_pool_t *dp, const char *name, ds_hold_flags_t flags, const void *tag, dsl_dataset_t **dsp) { return (dsl_dataset_own_impl(dp, name, flags, tag, B_TRUE, dsp)); } int dsl_dataset_own(dsl_pool_t *dp, const char *name, ds_hold_flags_t flags, const void *tag, dsl_dataset_t **dsp) { return (dsl_dataset_own_impl(dp, name, flags, tag, B_FALSE, dsp)); } /* * See the comment above dsl_pool_hold() for details. In summary, a long * hold is used to prevent destruction of a dataset while the pool hold * is dropped, allowing other concurrent operations (e.g. spa_sync()). * * The dataset and pool must be held when this function is called. After it * is called, the pool hold may be released while the dataset is still held * and accessed. */ void dsl_dataset_long_hold(dsl_dataset_t *ds, const void *tag) { ASSERT(dsl_pool_config_held(ds->ds_dir->dd_pool)); (void) zfs_refcount_add(&ds->ds_longholds, tag); } void dsl_dataset_long_rele(dsl_dataset_t *ds, const void *tag) { (void) zfs_refcount_remove(&ds->ds_longholds, tag); } /* Return B_TRUE if there are any long holds on this dataset. */ boolean_t dsl_dataset_long_held(dsl_dataset_t *ds) { return (!zfs_refcount_is_zero(&ds->ds_longholds)); } void dsl_dataset_name(dsl_dataset_t *ds, char *name) { if (ds == NULL) { (void) strlcpy(name, "mos", ZFS_MAX_DATASET_NAME_LEN); } else { dsl_dir_name(ds->ds_dir, name); VERIFY0(dsl_dataset_get_snapname(ds)); if (ds->ds_snapname[0]) { VERIFY3U(strlcat(name, "@", ZFS_MAX_DATASET_NAME_LEN), <, ZFS_MAX_DATASET_NAME_LEN); /* * We use a "recursive" mutex so that we * can call dprintf_ds() with ds_lock held. */ if (!MUTEX_HELD(&ds->ds_lock)) { mutex_enter(&ds->ds_lock); VERIFY3U(strlcat(name, ds->ds_snapname, ZFS_MAX_DATASET_NAME_LEN), <, ZFS_MAX_DATASET_NAME_LEN); mutex_exit(&ds->ds_lock); } else { VERIFY3U(strlcat(name, ds->ds_snapname, ZFS_MAX_DATASET_NAME_LEN), <, ZFS_MAX_DATASET_NAME_LEN); } } } } int dsl_dataset_namelen(dsl_dataset_t *ds) { VERIFY0(dsl_dataset_get_snapname(ds)); mutex_enter(&ds->ds_lock); int len = strlen(ds->ds_snapname); mutex_exit(&ds->ds_lock); /* add '@' if ds is a snap */ if (len > 0) len++; len += dsl_dir_namelen(ds->ds_dir); return (len); } void dsl_dataset_rele(dsl_dataset_t *ds, const void *tag) { dmu_buf_rele(ds->ds_dbuf, tag); } void dsl_dataset_remove_key_mapping(dsl_dataset_t *ds) { dsl_dir_t *dd = ds->ds_dir; if (dd == NULL || dd->dd_crypto_obj == 0) return; (void) spa_keystore_remove_mapping(dd->dd_pool->dp_spa, ds->ds_object, ds); } void dsl_dataset_rele_flags(dsl_dataset_t *ds, ds_hold_flags_t flags, const void *tag) { if (flags & DS_HOLD_FLAG_DECRYPT) dsl_dataset_remove_key_mapping(ds); dsl_dataset_rele(ds, tag); } void dsl_dataset_disown(dsl_dataset_t *ds, ds_hold_flags_t flags, const void *tag) { ASSERT3P(ds->ds_owner, ==, tag); ASSERT(ds->ds_dbuf != NULL); mutex_enter(&ds->ds_lock); ds->ds_owner = NULL; mutex_exit(&ds->ds_lock); dsl_dataset_long_rele(ds, tag); dsl_dataset_rele_flags(ds, flags, tag); } boolean_t dsl_dataset_tryown(dsl_dataset_t *ds, const void *tag, boolean_t override) { boolean_t gotit = FALSE; ASSERT(dsl_pool_config_held(ds->ds_dir->dd_pool)); mutex_enter(&ds->ds_lock); if (ds->ds_owner == NULL && (override || !(DS_IS_INCONSISTENT(ds) || (dsl_dataset_feature_is_active(ds, SPA_FEATURE_REDACTED_DATASETS) && !zfs_allow_redacted_dataset_mount)))) { ds->ds_owner = tag; dsl_dataset_long_hold(ds, tag); gotit = TRUE; } mutex_exit(&ds->ds_lock); return (gotit); } boolean_t dsl_dataset_has_owner(dsl_dataset_t *ds) { boolean_t rv; mutex_enter(&ds->ds_lock); rv = (ds->ds_owner != NULL); mutex_exit(&ds->ds_lock); return (rv); } static boolean_t zfeature_active(spa_feature_t f, void *arg) { switch (spa_feature_table[f].fi_type) { case ZFEATURE_TYPE_BOOLEAN: { boolean_t val = (boolean_t)(uintptr_t)arg; ASSERT(val == B_FALSE || val == B_TRUE); return (val); } case ZFEATURE_TYPE_UINT64_ARRAY: /* * In this case, arg is a uint64_t array. The feature is active * if the array is non-null. */ return (arg != NULL); default: panic("Invalid zfeature type %d", spa_feature_table[f].fi_type); return (B_FALSE); } } boolean_t dsl_dataset_feature_is_active(dsl_dataset_t *ds, spa_feature_t f) { return (zfeature_active(f, ds->ds_feature[f])); } /* * The buffers passed out by this function are references to internal buffers; * they should not be freed by callers of this function, and they should not be * used after the dataset has been released. */ boolean_t dsl_dataset_get_uint64_array_feature(dsl_dataset_t *ds, spa_feature_t f, uint64_t *outlength, uint64_t **outp) { VERIFY(spa_feature_table[f].fi_type & ZFEATURE_TYPE_UINT64_ARRAY); if (!dsl_dataset_feature_is_active(ds, f)) { return (B_FALSE); } struct feature_type_uint64_array_arg *ftuaa = ds->ds_feature[f]; *outp = ftuaa->array; *outlength = ftuaa->length; return (B_TRUE); } void dsl_dataset_activate_feature(uint64_t dsobj, spa_feature_t f, void *arg, dmu_tx_t *tx) { spa_t *spa = dmu_tx_pool(tx)->dp_spa; objset_t *mos = dmu_tx_pool(tx)->dp_meta_objset; uint64_t zero = 0; VERIFY(spa_feature_table[f].fi_flags & ZFEATURE_FLAG_PER_DATASET); spa_feature_incr(spa, f, tx); dmu_object_zapify(mos, dsobj, DMU_OT_DSL_DATASET, tx); switch (spa_feature_table[f].fi_type) { case ZFEATURE_TYPE_BOOLEAN: ASSERT3S((boolean_t)(uintptr_t)arg, ==, B_TRUE); VERIFY0(zap_add(mos, dsobj, spa_feature_table[f].fi_guid, sizeof (zero), 1, &zero, tx)); break; case ZFEATURE_TYPE_UINT64_ARRAY: { struct feature_type_uint64_array_arg *ftuaa = arg; VERIFY0(zap_add(mos, dsobj, spa_feature_table[f].fi_guid, sizeof (uint64_t), ftuaa->length, ftuaa->array, tx)); break; } default: panic("Invalid zfeature type %d", spa_feature_table[f].fi_type); } } static void dsl_dataset_deactivate_feature_impl(dsl_dataset_t *ds, spa_feature_t f, dmu_tx_t *tx) { spa_t *spa = dmu_tx_pool(tx)->dp_spa; objset_t *mos = dmu_tx_pool(tx)->dp_meta_objset; uint64_t dsobj = ds->ds_object; VERIFY(spa_feature_table[f].fi_flags & ZFEATURE_FLAG_PER_DATASET); VERIFY0(zap_remove(mos, dsobj, spa_feature_table[f].fi_guid, tx)); spa_feature_decr(spa, f, tx); ds->ds_feature[f] = NULL; } void dsl_dataset_deactivate_feature(dsl_dataset_t *ds, spa_feature_t f, dmu_tx_t *tx) { unload_zfeature(ds, f); dsl_dataset_deactivate_feature_impl(ds, f, tx); } uint64_t dsl_dataset_create_sync_dd(dsl_dir_t *dd, dsl_dataset_t *origin, dsl_crypto_params_t *dcp, uint64_t flags, dmu_tx_t *tx) { dsl_pool_t *dp = dd->dd_pool; dmu_buf_t *dbuf; dsl_dataset_phys_t *dsphys; uint64_t dsobj; objset_t *mos = dp->dp_meta_objset; if (origin == NULL) origin = dp->dp_origin_snap; ASSERT(origin == NULL || origin->ds_dir->dd_pool == dp); ASSERT(origin == NULL || dsl_dataset_phys(origin)->ds_num_children > 0); ASSERT(dmu_tx_is_syncing(tx)); ASSERT(dsl_dir_phys(dd)->dd_head_dataset_obj == 0); dsobj = dmu_object_alloc(mos, DMU_OT_DSL_DATASET, 0, DMU_OT_DSL_DATASET, sizeof (dsl_dataset_phys_t), tx); VERIFY0(dmu_bonus_hold(mos, dsobj, FTAG, &dbuf)); dmu_buf_will_dirty(dbuf, tx); dsphys = dbuf->db_data; memset(dsphys, 0, sizeof (dsl_dataset_phys_t)); dsphys->ds_dir_obj = dd->dd_object; dsphys->ds_flags = flags; dsphys->ds_fsid_guid = unique_create(); (void) random_get_pseudo_bytes((void*)&dsphys->ds_guid, sizeof (dsphys->ds_guid)); dsphys->ds_snapnames_zapobj = zap_create_norm(mos, U8_TEXTPREP_TOUPPER, DMU_OT_DSL_DS_SNAP_MAP, DMU_OT_NONE, 0, tx); dsphys->ds_creation_time = gethrestime_sec(); dsphys->ds_creation_txg = tx->tx_txg == TXG_INITIAL ? 1 : tx->tx_txg; if (origin == NULL) { dsphys->ds_deadlist_obj = dsl_deadlist_alloc(mos, tx); } else { dsl_dataset_t *ohds; /* head of the origin snapshot */ dsphys->ds_prev_snap_obj = origin->ds_object; dsphys->ds_prev_snap_txg = dsl_dataset_phys(origin)->ds_creation_txg; dsphys->ds_referenced_bytes = dsl_dataset_phys(origin)->ds_referenced_bytes; dsphys->ds_compressed_bytes = dsl_dataset_phys(origin)->ds_compressed_bytes; dsphys->ds_uncompressed_bytes = dsl_dataset_phys(origin)->ds_uncompressed_bytes; rrw_enter(&origin->ds_bp_rwlock, RW_READER, FTAG); dsphys->ds_bp = dsl_dataset_phys(origin)->ds_bp; rrw_exit(&origin->ds_bp_rwlock, FTAG); /* * Inherit flags that describe the dataset's contents * (INCONSISTENT) or properties (Case Insensitive). */ dsphys->ds_flags |= dsl_dataset_phys(origin)->ds_flags & (DS_FLAG_INCONSISTENT | DS_FLAG_CI_DATASET); for (spa_feature_t f = 0; f < SPA_FEATURES; f++) { if (zfeature_active(f, origin->ds_feature[f])) { dsl_dataset_activate_feature(dsobj, f, origin->ds_feature[f], tx); } } dmu_buf_will_dirty(origin->ds_dbuf, tx); dsl_dataset_phys(origin)->ds_num_children++; VERIFY0(dsl_dataset_hold_obj(dp, dsl_dir_phys(origin->ds_dir)->dd_head_dataset_obj, FTAG, &ohds)); dsphys->ds_deadlist_obj = dsl_deadlist_clone(&ohds->ds_deadlist, dsphys->ds_prev_snap_txg, dsphys->ds_prev_snap_obj, tx); dsl_dataset_rele(ohds, FTAG); if (spa_version(dp->dp_spa) >= SPA_VERSION_NEXT_CLONES) { if (dsl_dataset_phys(origin)->ds_next_clones_obj == 0) { dsl_dataset_phys(origin)->ds_next_clones_obj = zap_create(mos, DMU_OT_NEXT_CLONES, DMU_OT_NONE, 0, tx); } VERIFY0(zap_add_int(mos, dsl_dataset_phys(origin)->ds_next_clones_obj, dsobj, tx)); } dmu_buf_will_dirty(dd->dd_dbuf, tx); dsl_dir_phys(dd)->dd_origin_obj = origin->ds_object; if (spa_version(dp->dp_spa) >= SPA_VERSION_DIR_CLONES) { if (dsl_dir_phys(origin->ds_dir)->dd_clones == 0) { dmu_buf_will_dirty(origin->ds_dir->dd_dbuf, tx); dsl_dir_phys(origin->ds_dir)->dd_clones = zap_create(mos, DMU_OT_DSL_CLONES, DMU_OT_NONE, 0, tx); } VERIFY0(zap_add_int(mos, dsl_dir_phys(origin->ds_dir)->dd_clones, dsobj, tx)); } } /* handle encryption */ dsl_dataset_create_crypt_sync(dsobj, dd, origin, dcp, tx); if (spa_version(dp->dp_spa) >= SPA_VERSION_UNIQUE_ACCURATE) dsphys->ds_flags |= DS_FLAG_UNIQUE_ACCURATE; dmu_buf_rele(dbuf, FTAG); dmu_buf_will_dirty(dd->dd_dbuf, tx); dsl_dir_phys(dd)->dd_head_dataset_obj = dsobj; return (dsobj); } static void dsl_dataset_zero_zil(dsl_dataset_t *ds, dmu_tx_t *tx) { objset_t *os; VERIFY0(dmu_objset_from_ds(ds, &os)); if (memcmp(&os->os_zil_header, &zero_zil, sizeof (zero_zil)) != 0) { dsl_pool_t *dp = ds->ds_dir->dd_pool; zio_t *zio; memset(&os->os_zil_header, 0, sizeof (os->os_zil_header)); if (os->os_encrypted) os->os_next_write_raw[tx->tx_txg & TXG_MASK] = B_TRUE; zio = zio_root(dp->dp_spa, NULL, NULL, ZIO_FLAG_MUSTSUCCEED); dsl_dataset_sync(ds, zio, tx); VERIFY0(zio_wait(zio)); dsl_dataset_sync_done(ds, tx); } } uint64_t dsl_dataset_create_sync(dsl_dir_t *pdd, const char *lastname, dsl_dataset_t *origin, uint64_t flags, cred_t *cr, dsl_crypto_params_t *dcp, dmu_tx_t *tx) { dsl_pool_t *dp = pdd->dd_pool; uint64_t dsobj, ddobj; dsl_dir_t *dd; ASSERT(dmu_tx_is_syncing(tx)); ASSERT(lastname[0] != '@'); /* * Filesystems will eventually have their origin set to dp_origin_snap, * but that's taken care of in dsl_dataset_create_sync_dd. When * creating a filesystem, this function is called with origin equal to * NULL. */ if (origin != NULL) ASSERT3P(origin, !=, dp->dp_origin_snap); ddobj = dsl_dir_create_sync(dp, pdd, lastname, tx); VERIFY0(dsl_dir_hold_obj(dp, ddobj, lastname, FTAG, &dd)); dsobj = dsl_dataset_create_sync_dd(dd, origin, dcp, flags & ~DS_CREATE_FLAG_NODIRTY, tx); dsl_deleg_set_create_perms(dd, tx, cr); /* * If we are creating a clone and the livelist feature is enabled, * add the entry DD_FIELD_LIVELIST to ZAP. */ if (origin != NULL && spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_LIVELIST)) { objset_t *mos = dd->dd_pool->dp_meta_objset; dsl_dir_zapify(dd, tx); uint64_t obj = dsl_deadlist_alloc(mos, tx); VERIFY0(zap_add(mos, dd->dd_object, DD_FIELD_LIVELIST, sizeof (uint64_t), 1, &obj, tx)); spa_feature_incr(dp->dp_spa, SPA_FEATURE_LIVELIST, tx); } /* * Since we're creating a new node we know it's a leaf, so we can * initialize the counts if the limit feature is active. */ if (spa_feature_is_active(dp->dp_spa, SPA_FEATURE_FS_SS_LIMIT)) { uint64_t cnt = 0; objset_t *os = dd->dd_pool->dp_meta_objset; dsl_dir_zapify(dd, tx); VERIFY0(zap_add(os, dd->dd_object, DD_FIELD_FILESYSTEM_COUNT, sizeof (cnt), 1, &cnt, tx)); VERIFY0(zap_add(os, dd->dd_object, DD_FIELD_SNAPSHOT_COUNT, sizeof (cnt), 1, &cnt, tx)); } dsl_dir_rele(dd, FTAG); /* * If we are creating a clone, make sure we zero out any stale * data from the origin snapshots zil header. */ if (origin != NULL && !(flags & DS_CREATE_FLAG_NODIRTY)) { dsl_dataset_t *ds; VERIFY0(dsl_dataset_hold_obj(dp, dsobj, FTAG, &ds)); dsl_dataset_zero_zil(ds, tx); dsl_dataset_rele(ds, FTAG); } return (dsobj); } /* * The unique space in the head dataset can be calculated by subtracting * the space used in the most recent snapshot, that is still being used * in this file system, from the space currently in use. To figure out * the space in the most recent snapshot still in use, we need to take * the total space used in the snapshot and subtract out the space that * has been freed up since the snapshot was taken. */ void dsl_dataset_recalc_head_uniq(dsl_dataset_t *ds) { uint64_t mrs_used; uint64_t dlused, dlcomp, dluncomp; ASSERT(!ds->ds_is_snapshot); if (dsl_dataset_phys(ds)->ds_prev_snap_obj != 0) mrs_used = dsl_dataset_phys(ds->ds_prev)->ds_referenced_bytes; else mrs_used = 0; dsl_deadlist_space(&ds->ds_deadlist, &dlused, &dlcomp, &dluncomp); ASSERT3U(dlused, <=, mrs_used); dsl_dataset_phys(ds)->ds_unique_bytes = dsl_dataset_phys(ds)->ds_referenced_bytes - (mrs_used - dlused); if (spa_version(ds->ds_dir->dd_pool->dp_spa) >= SPA_VERSION_UNIQUE_ACCURATE) dsl_dataset_phys(ds)->ds_flags |= DS_FLAG_UNIQUE_ACCURATE; } void dsl_dataset_remove_from_next_clones(dsl_dataset_t *ds, uint64_t obj, dmu_tx_t *tx) { objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset; uint64_t count __maybe_unused; int err; ASSERT(dsl_dataset_phys(ds)->ds_num_children >= 2); err = zap_remove_int(mos, dsl_dataset_phys(ds)->ds_next_clones_obj, obj, tx); /* * The err should not be ENOENT, but a bug in a previous version * of the code could cause upgrade_clones_cb() to not set * ds_next_snap_obj when it should, leading to a missing entry. * If we knew that the pool was created after * SPA_VERSION_NEXT_CLONES, we could assert that it isn't * ENOENT. However, at least we can check that we don't have * too many entries in the next_clones_obj even after failing to * remove this one. */ if (err != ENOENT) VERIFY0(err); ASSERT0(zap_count(mos, dsl_dataset_phys(ds)->ds_next_clones_obj, &count)); ASSERT3U(count, <=, dsl_dataset_phys(ds)->ds_num_children - 2); } blkptr_t * dsl_dataset_get_blkptr(dsl_dataset_t *ds) { return (&dsl_dataset_phys(ds)->ds_bp); } spa_t * dsl_dataset_get_spa(dsl_dataset_t *ds) { return (ds->ds_dir->dd_pool->dp_spa); } void dsl_dataset_dirty(dsl_dataset_t *ds, dmu_tx_t *tx) { dsl_pool_t *dp; if (ds == NULL) /* this is the meta-objset */ return; ASSERT(ds->ds_objset != NULL); if (dsl_dataset_phys(ds)->ds_next_snap_obj != 0) panic("dirtying snapshot!"); /* Must not dirty a dataset in the same txg where it got snapshotted. */ ASSERT3U(tx->tx_txg, >, dsl_dataset_phys(ds)->ds_prev_snap_txg); dp = ds->ds_dir->dd_pool; if (txg_list_add(&dp->dp_dirty_datasets, ds, tx->tx_txg)) { objset_t *os = ds->ds_objset; /* up the hold count until we can be written out */ dmu_buf_add_ref(ds->ds_dbuf, ds); /* if this dataset is encrypted, grab a reference to the DCK */ if (ds->ds_dir->dd_crypto_obj != 0 && !os->os_raw_receive && !os->os_next_write_raw[tx->tx_txg & TXG_MASK]) { ASSERT3P(ds->ds_key_mapping, !=, NULL); key_mapping_add_ref(ds->ds_key_mapping, ds); } } } static int dsl_dataset_snapshot_reserve_space(dsl_dataset_t *ds, dmu_tx_t *tx) { uint64_t asize; if (!dmu_tx_is_syncing(tx)) return (0); /* * If there's an fs-only reservation, any blocks that might become * owned by the snapshot dataset must be accommodated by space * outside of the reservation. */ ASSERT(ds->ds_reserved == 0 || DS_UNIQUE_IS_ACCURATE(ds)); asize = MIN(dsl_dataset_phys(ds)->ds_unique_bytes, ds->ds_reserved); if (asize > dsl_dir_space_available(ds->ds_dir, NULL, 0, TRUE)) return (SET_ERROR(ENOSPC)); /* * Propagate any reserved space for this snapshot to other * snapshot checks in this sync group. */ if (asize > 0) dsl_dir_willuse_space(ds->ds_dir, asize, tx); return (0); } int dsl_dataset_snapshot_check_impl(dsl_dataset_t *ds, const char *snapname, dmu_tx_t *tx, boolean_t recv, uint64_t cnt, cred_t *cr, proc_t *proc) { int error; uint64_t value; ds->ds_trysnap_txg = tx->tx_txg; if (!dmu_tx_is_syncing(tx)) return (0); /* * We don't allow multiple snapshots of the same txg. If there * is already one, try again. */ if (dsl_dataset_phys(ds)->ds_prev_snap_txg >= tx->tx_txg) return (SET_ERROR(EAGAIN)); /* * Check for conflicting snapshot name. */ error = dsl_dataset_snap_lookup(ds, snapname, &value); if (error == 0) return (SET_ERROR(EEXIST)); if (error != ENOENT) return (error); /* * We don't allow taking snapshots of inconsistent datasets, such as * those into which we are currently receiving. However, if we are * creating this snapshot as part of a receive, this check will be * executed atomically with respect to the completion of the receive * itself but prior to the clearing of DS_FLAG_INCONSISTENT; in this * case we ignore this, knowing it will be fixed up for us shortly in * dmu_recv_end_sync(). */ if (!recv && DS_IS_INCONSISTENT(ds)) return (SET_ERROR(EBUSY)); /* * Skip the check for temporary snapshots or if we have already checked * the counts in dsl_dataset_snapshot_check. This means we really only * check the count here when we're receiving a stream. */ if (cnt != 0 && cr != NULL) { error = dsl_fs_ss_limit_check(ds->ds_dir, cnt, ZFS_PROP_SNAPSHOT_LIMIT, NULL, cr, proc); if (error != 0) return (error); } error = dsl_dataset_snapshot_reserve_space(ds, tx); if (error != 0) return (error); return (0); } int dsl_dataset_snapshot_check(void *arg, dmu_tx_t *tx) { dsl_dataset_snapshot_arg_t *ddsa = arg; dsl_pool_t *dp = dmu_tx_pool(tx); nvpair_t *pair; int rv = 0; /* * Pre-compute how many total new snapshots will be created for each * level in the tree and below. This is needed for validating the * snapshot limit when either taking a recursive snapshot or when * taking multiple snapshots. * * The problem is that the counts are not actually adjusted when * we are checking, only when we finally sync. For a single snapshot, * this is easy, the count will increase by 1 at each node up the tree, * but its more complicated for the recursive/multiple snapshot case. * * The dsl_fs_ss_limit_check function does recursively check the count * at each level up the tree but since it is validating each snapshot * independently we need to be sure that we are validating the complete * count for the entire set of snapshots. We do this by rolling up the * counts for each component of the name into an nvlist and then * checking each of those cases with the aggregated count. * * This approach properly handles not only the recursive snapshot * case (where we get all of those on the ddsa_snaps list) but also * the sibling case (e.g. snapshot a/b and a/c so that we will also * validate the limit on 'a' using a count of 2). * * We validate the snapshot names in the third loop and only report * name errors once. */ if (dmu_tx_is_syncing(tx)) { char *nm; nvlist_t *cnt_track = NULL; cnt_track = fnvlist_alloc(); nm = kmem_alloc(MAXPATHLEN, KM_SLEEP); /* Rollup aggregated counts into the cnt_track list */ for (pair = nvlist_next_nvpair(ddsa->ddsa_snaps, NULL); pair != NULL; pair = nvlist_next_nvpair(ddsa->ddsa_snaps, pair)) { char *pdelim; uint64_t val; (void) strlcpy(nm, nvpair_name(pair), MAXPATHLEN); pdelim = strchr(nm, '@'); if (pdelim == NULL) continue; *pdelim = '\0'; do { if (nvlist_lookup_uint64(cnt_track, nm, &val) == 0) { /* update existing entry */ fnvlist_add_uint64(cnt_track, nm, val + 1); } else { /* add to list */ fnvlist_add_uint64(cnt_track, nm, 1); } pdelim = strrchr(nm, '/'); if (pdelim != NULL) *pdelim = '\0'; } while (pdelim != NULL); } kmem_free(nm, MAXPATHLEN); /* Check aggregated counts at each level */ for (pair = nvlist_next_nvpair(cnt_track, NULL); pair != NULL; pair = nvlist_next_nvpair(cnt_track, pair)) { int error = 0; const char *name; uint64_t cnt = 0; dsl_dataset_t *ds; name = nvpair_name(pair); cnt = fnvpair_value_uint64(pair); ASSERT(cnt > 0); error = dsl_dataset_hold(dp, name, FTAG, &ds); if (error == 0) { error = dsl_fs_ss_limit_check(ds->ds_dir, cnt, ZFS_PROP_SNAPSHOT_LIMIT, NULL, ddsa->ddsa_cr, ddsa->ddsa_proc); dsl_dataset_rele(ds, FTAG); } if (error != 0) { if (ddsa->ddsa_errors != NULL) fnvlist_add_int32(ddsa->ddsa_errors, name, error); rv = error; /* only report one error for this check */ break; } } nvlist_free(cnt_track); } for (pair = nvlist_next_nvpair(ddsa->ddsa_snaps, NULL); pair != NULL; pair = nvlist_next_nvpair(ddsa->ddsa_snaps, pair)) { int error = 0; dsl_dataset_t *ds; const char *name, *atp = NULL; char dsname[ZFS_MAX_DATASET_NAME_LEN]; name = nvpair_name(pair); if (strlen(name) >= ZFS_MAX_DATASET_NAME_LEN) error = SET_ERROR(ENAMETOOLONG); if (error == 0) { atp = strchr(name, '@'); if (atp == NULL) error = SET_ERROR(EINVAL); if (error == 0) (void) strlcpy(dsname, name, atp - name + 1); } if (error == 0) error = dsl_dataset_hold(dp, dsname, FTAG, &ds); if (error == 0) { /* passing 0/NULL skips dsl_fs_ss_limit_check */ error = dsl_dataset_snapshot_check_impl(ds, atp + 1, tx, B_FALSE, 0, NULL, NULL); dsl_dataset_rele(ds, FTAG); } if (error != 0) { if (ddsa->ddsa_errors != NULL) { fnvlist_add_int32(ddsa->ddsa_errors, name, error); } rv = error; } } return (rv); } void dsl_dataset_snapshot_sync_impl(dsl_dataset_t *ds, const char *snapname, dmu_tx_t *tx) { dsl_pool_t *dp = ds->ds_dir->dd_pool; dmu_buf_t *dbuf; dsl_dataset_phys_t *dsphys; uint64_t dsobj, crtxg; objset_t *mos = dp->dp_meta_objset; objset_t *os __maybe_unused; ASSERT(RRW_WRITE_HELD(&dp->dp_config_rwlock)); /* * If we are on an old pool, the zil must not be active, in which * case it will be zeroed. Usually zil_suspend() accomplishes this. */ ASSERT(spa_version(dmu_tx_pool(tx)->dp_spa) >= SPA_VERSION_FAST_SNAP || dmu_objset_from_ds(ds, &os) != 0 || memcmp(&os->os_phys->os_zil_header, &zero_zil, sizeof (zero_zil)) == 0); /* Should not snapshot a dirty dataset. */ ASSERT(!txg_list_member(&ds->ds_dir->dd_pool->dp_dirty_datasets, ds, tx->tx_txg)); dsl_fs_ss_count_adjust(ds->ds_dir, 1, DD_FIELD_SNAPSHOT_COUNT, tx); /* * The origin's ds_creation_txg has to be < TXG_INITIAL */ if (strcmp(snapname, ORIGIN_DIR_NAME) == 0) crtxg = 1; else crtxg = tx->tx_txg; dsobj = dmu_object_alloc(mos, DMU_OT_DSL_DATASET, 0, DMU_OT_DSL_DATASET, sizeof (dsl_dataset_phys_t), tx); VERIFY0(dmu_bonus_hold(mos, dsobj, FTAG, &dbuf)); dmu_buf_will_dirty(dbuf, tx); dsphys = dbuf->db_data; memset(dsphys, 0, sizeof (dsl_dataset_phys_t)); dsphys->ds_dir_obj = ds->ds_dir->dd_object; dsphys->ds_fsid_guid = unique_create(); (void) random_get_pseudo_bytes((void*)&dsphys->ds_guid, sizeof (dsphys->ds_guid)); dsphys->ds_prev_snap_obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; dsphys->ds_prev_snap_txg = dsl_dataset_phys(ds)->ds_prev_snap_txg; dsphys->ds_next_snap_obj = ds->ds_object; dsphys->ds_num_children = 1; dsphys->ds_creation_time = gethrestime_sec(); dsphys->ds_creation_txg = crtxg; dsphys->ds_deadlist_obj = dsl_dataset_phys(ds)->ds_deadlist_obj; dsphys->ds_referenced_bytes = dsl_dataset_phys(ds)->ds_referenced_bytes; dsphys->ds_compressed_bytes = dsl_dataset_phys(ds)->ds_compressed_bytes; dsphys->ds_uncompressed_bytes = dsl_dataset_phys(ds)->ds_uncompressed_bytes; dsphys->ds_flags = dsl_dataset_phys(ds)->ds_flags; rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG); dsphys->ds_bp = dsl_dataset_phys(ds)->ds_bp; rrw_exit(&ds->ds_bp_rwlock, FTAG); dmu_buf_rele(dbuf, FTAG); for (spa_feature_t f = 0; f < SPA_FEATURES; f++) { if (zfeature_active(f, ds->ds_feature[f])) { dsl_dataset_activate_feature(dsobj, f, ds->ds_feature[f], tx); } } ASSERT3U(ds->ds_prev != 0, ==, dsl_dataset_phys(ds)->ds_prev_snap_obj != 0); if (ds->ds_prev) { uint64_t next_clones_obj = dsl_dataset_phys(ds->ds_prev)->ds_next_clones_obj; ASSERT(dsl_dataset_phys(ds->ds_prev)->ds_next_snap_obj == ds->ds_object || dsl_dataset_phys(ds->ds_prev)->ds_num_children > 1); if (dsl_dataset_phys(ds->ds_prev)->ds_next_snap_obj == ds->ds_object) { dmu_buf_will_dirty(ds->ds_prev->ds_dbuf, tx); ASSERT3U(dsl_dataset_phys(ds)->ds_prev_snap_txg, ==, dsl_dataset_phys(ds->ds_prev)->ds_creation_txg); dsl_dataset_phys(ds->ds_prev)->ds_next_snap_obj = dsobj; } else if (next_clones_obj != 0) { dsl_dataset_remove_from_next_clones(ds->ds_prev, dsphys->ds_next_snap_obj, tx); VERIFY0(zap_add_int(mos, next_clones_obj, dsobj, tx)); } } /* * If we have a reference-reservation on this dataset, we will * need to increase the amount of refreservation being charged * since our unique space is going to zero. */ if (ds->ds_reserved) { int64_t delta; ASSERT(DS_UNIQUE_IS_ACCURATE(ds)); delta = MIN(dsl_dataset_phys(ds)->ds_unique_bytes, ds->ds_reserved); dsl_dir_diduse_space(ds->ds_dir, DD_USED_REFRSRV, delta, 0, 0, tx); } dmu_buf_will_dirty(ds->ds_dbuf, tx); dsl_dataset_phys(ds)->ds_deadlist_obj = dsl_deadlist_clone(&ds->ds_deadlist, UINT64_MAX, dsl_dataset_phys(ds)->ds_prev_snap_obj, tx); dsl_deadlist_close(&ds->ds_deadlist); dsl_deadlist_open(&ds->ds_deadlist, mos, dsl_dataset_phys(ds)->ds_deadlist_obj); dsl_deadlist_add_key(&ds->ds_deadlist, dsl_dataset_phys(ds)->ds_prev_snap_txg, tx); dsl_bookmark_snapshotted(ds, tx); if (dsl_dataset_remap_deadlist_exists(ds)) { uint64_t remap_deadlist_obj = dsl_dataset_get_remap_deadlist_object(ds); /* * Move the remap_deadlist to the snapshot. The head * will create a new remap deadlist on demand, from * dsl_dataset_block_remapped(). */ dsl_dataset_unset_remap_deadlist_object(ds, tx); dsl_deadlist_close(&ds->ds_remap_deadlist); dmu_object_zapify(mos, dsobj, DMU_OT_DSL_DATASET, tx); VERIFY0(zap_add(mos, dsobj, DS_FIELD_REMAP_DEADLIST, sizeof (remap_deadlist_obj), 1, &remap_deadlist_obj, tx)); } /* * Create a ivset guid for this snapshot if the dataset is * encrypted. This may be overridden by a raw receive. A * previous implementation of this code did not have this * field as part of the on-disk format for ZFS encryption * (see errata #4). As part of the remediation for this * issue, we ask the user to enable the bookmark_v2 feature * which is now a dependency of the encryption feature. We * use this as a heuristic to determine when the user has * elected to correct any datasets created with the old code. * As a result, we only do this step if the bookmark_v2 * feature is enabled, which limits the number of states a * given pool / dataset can be in with regards to terms of * correcting the issue. */ if (ds->ds_dir->dd_crypto_obj != 0 && spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_BOOKMARK_V2)) { uint64_t ivset_guid = unique_create(); dmu_object_zapify(mos, dsobj, DMU_OT_DSL_DATASET, tx); VERIFY0(zap_add(mos, dsobj, DS_FIELD_IVSET_GUID, sizeof (ivset_guid), 1, &ivset_guid, tx)); } ASSERT3U(dsl_dataset_phys(ds)->ds_prev_snap_txg, <, tx->tx_txg); dsl_dataset_phys(ds)->ds_prev_snap_obj = dsobj; dsl_dataset_phys(ds)->ds_prev_snap_txg = crtxg; dsl_dataset_phys(ds)->ds_unique_bytes = 0; if (spa_version(dp->dp_spa) >= SPA_VERSION_UNIQUE_ACCURATE) dsl_dataset_phys(ds)->ds_flags |= DS_FLAG_UNIQUE_ACCURATE; VERIFY0(zap_add(mos, dsl_dataset_phys(ds)->ds_snapnames_zapobj, snapname, 8, 1, &dsobj, tx)); if (ds->ds_prev) dsl_dataset_rele(ds->ds_prev, ds); VERIFY0(dsl_dataset_hold_obj(dp, dsl_dataset_phys(ds)->ds_prev_snap_obj, ds, &ds->ds_prev)); dsl_scan_ds_snapshotted(ds, tx); dsl_dir_snap_cmtime_update(ds->ds_dir, tx); if (zfs_snapshot_history_enabled) spa_history_log_internal_ds(ds->ds_prev, "snapshot", tx, " "); } void dsl_dataset_snapshot_sync(void *arg, dmu_tx_t *tx) { dsl_dataset_snapshot_arg_t *ddsa = arg; dsl_pool_t *dp = dmu_tx_pool(tx); nvpair_t *pair; for (pair = nvlist_next_nvpair(ddsa->ddsa_snaps, NULL); pair != NULL; pair = nvlist_next_nvpair(ddsa->ddsa_snaps, pair)) { dsl_dataset_t *ds; const char *name, *atp; char dsname[ZFS_MAX_DATASET_NAME_LEN]; name = nvpair_name(pair); atp = strchr(name, '@'); (void) strlcpy(dsname, name, atp - name + 1); VERIFY0(dsl_dataset_hold(dp, dsname, FTAG, &ds)); dsl_dataset_snapshot_sync_impl(ds, atp + 1, tx); if (ddsa->ddsa_props != NULL) { dsl_props_set_sync_impl(ds->ds_prev, ZPROP_SRC_LOCAL, ddsa->ddsa_props, tx); } dsl_dataset_rele(ds, FTAG); } } /* * The snapshots must all be in the same pool. * All-or-nothing: if there are any failures, nothing will be modified. */ int dsl_dataset_snapshot(nvlist_t *snaps, nvlist_t *props, nvlist_t *errors) { dsl_dataset_snapshot_arg_t ddsa; nvpair_t *pair; boolean_t needsuspend; int error; spa_t *spa; const char *firstname; nvlist_t *suspended = NULL; pair = nvlist_next_nvpair(snaps, NULL); if (pair == NULL) return (0); firstname = nvpair_name(pair); error = spa_open(firstname, &spa, FTAG); if (error != 0) return (error); needsuspend = (spa_version(spa) < SPA_VERSION_FAST_SNAP); spa_close(spa, FTAG); if (needsuspend) { suspended = fnvlist_alloc(); for (pair = nvlist_next_nvpair(snaps, NULL); pair != NULL; pair = nvlist_next_nvpair(snaps, pair)) { char fsname[ZFS_MAX_DATASET_NAME_LEN]; const char *snapname = nvpair_name(pair); const char *atp; void *cookie; atp = strchr(snapname, '@'); if (atp == NULL) { error = SET_ERROR(EINVAL); break; } (void) strlcpy(fsname, snapname, atp - snapname + 1); error = zil_suspend(fsname, &cookie); if (error != 0) break; fnvlist_add_uint64(suspended, fsname, (uintptr_t)cookie); } } ddsa.ddsa_snaps = snaps; ddsa.ddsa_props = props; ddsa.ddsa_errors = errors; ddsa.ddsa_cr = CRED(); ddsa.ddsa_proc = curproc; if (error == 0) { error = dsl_sync_task(firstname, dsl_dataset_snapshot_check, dsl_dataset_snapshot_sync, &ddsa, fnvlist_num_pairs(snaps) * 3, ZFS_SPACE_CHECK_NORMAL); } if (suspended != NULL) { for (pair = nvlist_next_nvpair(suspended, NULL); pair != NULL; pair = nvlist_next_nvpair(suspended, pair)) { zil_resume((void *)(uintptr_t) fnvpair_value_uint64(pair)); } fnvlist_free(suspended); } if (error == 0) { for (pair = nvlist_next_nvpair(snaps, NULL); pair != NULL; pair = nvlist_next_nvpair(snaps, pair)) { zvol_create_minor(nvpair_name(pair)); } } return (error); } typedef struct dsl_dataset_snapshot_tmp_arg { const char *ddsta_fsname; const char *ddsta_snapname; minor_t ddsta_cleanup_minor; const char *ddsta_htag; } dsl_dataset_snapshot_tmp_arg_t; static int dsl_dataset_snapshot_tmp_check(void *arg, dmu_tx_t *tx) { dsl_dataset_snapshot_tmp_arg_t *ddsta = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *ds; int error; error = dsl_dataset_hold(dp, ddsta->ddsta_fsname, FTAG, &ds); if (error != 0) return (error); /* NULL cred means no limit check for tmp snapshot */ error = dsl_dataset_snapshot_check_impl(ds, ddsta->ddsta_snapname, tx, B_FALSE, 0, NULL, NULL); if (error != 0) { dsl_dataset_rele(ds, FTAG); return (error); } if (spa_version(dp->dp_spa) < SPA_VERSION_USERREFS) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(ENOTSUP)); } error = dsl_dataset_user_hold_check_one(NULL, ddsta->ddsta_htag, B_TRUE, tx); if (error != 0) { dsl_dataset_rele(ds, FTAG); return (error); } dsl_dataset_rele(ds, FTAG); return (0); } static void dsl_dataset_snapshot_tmp_sync(void *arg, dmu_tx_t *tx) { dsl_dataset_snapshot_tmp_arg_t *ddsta = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *ds = NULL; VERIFY0(dsl_dataset_hold(dp, ddsta->ddsta_fsname, FTAG, &ds)); dsl_dataset_snapshot_sync_impl(ds, ddsta->ddsta_snapname, tx); dsl_dataset_user_hold_sync_one(ds->ds_prev, ddsta->ddsta_htag, ddsta->ddsta_cleanup_minor, gethrestime_sec(), tx); dsl_destroy_snapshot_sync_impl(ds->ds_prev, B_TRUE, tx); dsl_dataset_rele(ds, FTAG); } int dsl_dataset_snapshot_tmp(const char *fsname, const char *snapname, minor_t cleanup_minor, const char *htag) { dsl_dataset_snapshot_tmp_arg_t ddsta; int error; spa_t *spa; boolean_t needsuspend; void *cookie; ddsta.ddsta_fsname = fsname; ddsta.ddsta_snapname = snapname; ddsta.ddsta_cleanup_minor = cleanup_minor; ddsta.ddsta_htag = htag; error = spa_open(fsname, &spa, FTAG); if (error != 0) return (error); needsuspend = (spa_version(spa) < SPA_VERSION_FAST_SNAP); spa_close(spa, FTAG); if (needsuspend) { error = zil_suspend(fsname, &cookie); if (error != 0) return (error); } error = dsl_sync_task(fsname, dsl_dataset_snapshot_tmp_check, dsl_dataset_snapshot_tmp_sync, &ddsta, 3, ZFS_SPACE_CHECK_RESERVED); if (needsuspend) zil_resume(cookie); return (error); } /* Nonblocking dataset sync. Assumes dataset:objset is always 1:1 */ void dsl_dataset_sync(dsl_dataset_t *ds, zio_t *rio, dmu_tx_t *tx) { ASSERT(dmu_tx_is_syncing(tx)); ASSERT(ds->ds_objset != NULL); ASSERT(dsl_dataset_phys(ds)->ds_next_snap_obj == 0); /* * in case we had to change ds_fsid_guid when we opened it, * sync it out now. */ dmu_buf_will_dirty(ds->ds_dbuf, tx); dsl_dataset_phys(ds)->ds_fsid_guid = ds->ds_fsid_guid; if (ds->ds_resume_bytes[tx->tx_txg & TXG_MASK] != 0) { VERIFY0(zap_update(tx->tx_pool->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_OBJECT, 8, 1, &ds->ds_resume_object[tx->tx_txg & TXG_MASK], tx)); VERIFY0(zap_update(tx->tx_pool->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_OFFSET, 8, 1, &ds->ds_resume_offset[tx->tx_txg & TXG_MASK], tx)); VERIFY0(zap_update(tx->tx_pool->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_BYTES, 8, 1, &ds->ds_resume_bytes[tx->tx_txg & TXG_MASK], tx)); ds->ds_resume_object[tx->tx_txg & TXG_MASK] = 0; ds->ds_resume_offset[tx->tx_txg & TXG_MASK] = 0; ds->ds_resume_bytes[tx->tx_txg & TXG_MASK] = 0; } dmu_objset_sync(ds->ds_objset, rio, tx); } /* * Check if the percentage of blocks shared between the clone and the * snapshot (as opposed to those that are clone only) is below a certain * threshold */ static boolean_t dsl_livelist_should_disable(dsl_dataset_t *ds) { uint64_t used, referenced; int percent_shared; used = dsl_dir_get_usedds(ds->ds_dir); referenced = dsl_get_referenced(ds); if (referenced == 0) return (B_FALSE); percent_shared = (100 * (referenced - used)) / referenced; if (percent_shared <= zfs_livelist_min_percent_shared) return (B_TRUE); return (B_FALSE); } /* * Check if it is possible to combine two livelist entries into one. * This is the case if the combined number of 'live' blkptrs (ALLOCs that * don't have a matching FREE) is under the maximum sublist size. * We check this by subtracting twice the total number of frees from the total * number of blkptrs. FREEs are counted twice because each FREE blkptr * will cancel out an ALLOC blkptr when the livelist is processed. */ static boolean_t dsl_livelist_should_condense(dsl_deadlist_entry_t *first, dsl_deadlist_entry_t *next) { uint64_t total_free = first->dle_bpobj.bpo_phys->bpo_num_freed + next->dle_bpobj.bpo_phys->bpo_num_freed; uint64_t total_entries = first->dle_bpobj.bpo_phys->bpo_num_blkptrs + next->dle_bpobj.bpo_phys->bpo_num_blkptrs; if ((total_entries - (2 * total_free)) < zfs_livelist_max_entries) return (B_TRUE); return (B_FALSE); } typedef struct try_condense_arg { spa_t *spa; dsl_dataset_t *ds; } try_condense_arg_t; /* * Iterate over the livelist entries, searching for a pair to condense. * A nonzero return value means stop, 0 means keep looking. */ static int dsl_livelist_try_condense(void *arg, dsl_deadlist_entry_t *first) { try_condense_arg_t *tca = arg; spa_t *spa = tca->spa; dsl_dataset_t *ds = tca->ds; dsl_deadlist_t *ll = &ds->ds_dir->dd_livelist; dsl_deadlist_entry_t *next; /* The condense thread has not yet been created at import */ if (spa->spa_livelist_condense_zthr == NULL) return (1); /* A condense is already in progress */ if (spa->spa_to_condense.ds != NULL) return (1); next = AVL_NEXT(&ll->dl_tree, &first->dle_node); /* The livelist has only one entry - don't condense it */ if (next == NULL) return (1); /* Next is the newest entry - don't condense it */ if (AVL_NEXT(&ll->dl_tree, &next->dle_node) == NULL) return (1); /* This pair is not ready to condense but keep looking */ if (!dsl_livelist_should_condense(first, next)) return (0); /* * Add a ref to prevent the dataset from being evicted while * the condense zthr or synctask are running. Ref will be * released at the end of the condense synctask */ dmu_buf_add_ref(ds->ds_dbuf, spa); spa->spa_to_condense.ds = ds; spa->spa_to_condense.first = first; spa->spa_to_condense.next = next; spa->spa_to_condense.syncing = B_FALSE; spa->spa_to_condense.cancelled = B_FALSE; zthr_wakeup(spa->spa_livelist_condense_zthr); return (1); } static void dsl_flush_pending_livelist(dsl_dataset_t *ds, dmu_tx_t *tx) { dsl_dir_t *dd = ds->ds_dir; spa_t *spa = ds->ds_dir->dd_pool->dp_spa; dsl_deadlist_entry_t *last = dsl_deadlist_last(&dd->dd_livelist); /* Check if we need to add a new sub-livelist */ if (last == NULL) { /* The livelist is empty */ dsl_deadlist_add_key(&dd->dd_livelist, tx->tx_txg - 1, tx); } else if (spa_sync_pass(spa) == 1) { /* * Check if the newest entry is full. If it is, make a new one. * We only do this once per sync because we could overfill a * sublist in one sync pass and don't want to add another entry * for a txg that is already represented. This ensures that * blkptrs born in the same txg are stored in the same sublist. */ bpobj_t bpobj = last->dle_bpobj; uint64_t all = bpobj.bpo_phys->bpo_num_blkptrs; uint64_t free = bpobj.bpo_phys->bpo_num_freed; uint64_t alloc = all - free; if (alloc > zfs_livelist_max_entries) { dsl_deadlist_add_key(&dd->dd_livelist, tx->tx_txg - 1, tx); } } /* Insert each entry into the on-disk livelist */ bplist_iterate(&dd->dd_pending_allocs, dsl_deadlist_insert_alloc_cb, &dd->dd_livelist, tx); bplist_iterate(&dd->dd_pending_frees, dsl_deadlist_insert_free_cb, &dd->dd_livelist, tx); /* Attempt to condense every pair of adjacent entries */ try_condense_arg_t arg = { .spa = spa, .ds = ds }; dsl_deadlist_iterate(&dd->dd_livelist, dsl_livelist_try_condense, &arg); } void dsl_dataset_sync_done(dsl_dataset_t *ds, dmu_tx_t *tx) { objset_t *os = ds->ds_objset; bplist_iterate(&ds->ds_pending_deadlist, dsl_deadlist_insert_alloc_cb, &ds->ds_deadlist, tx); if (dsl_deadlist_is_open(&ds->ds_dir->dd_livelist)) { dsl_flush_pending_livelist(ds, tx); if (dsl_livelist_should_disable(ds)) { dsl_dir_remove_livelist(ds->ds_dir, tx, B_TRUE); } } dsl_bookmark_sync_done(ds, tx); multilist_destroy(&os->os_synced_dnodes); if (os->os_encrypted) os->os_next_write_raw[tx->tx_txg & TXG_MASK] = B_FALSE; else ASSERT0(os->os_next_write_raw[tx->tx_txg & TXG_MASK]); for (spa_feature_t f = 0; f < SPA_FEATURES; f++) { if (zfeature_active(f, ds->ds_feature_activation[f])) { if (zfeature_active(f, ds->ds_feature[f])) continue; dsl_dataset_activate_feature(ds->ds_object, f, ds->ds_feature_activation[f], tx); ds->ds_feature[f] = ds->ds_feature_activation[f]; } } ASSERT(!dmu_objset_is_dirty(os, dmu_tx_get_txg(tx))); } int get_clones_stat_impl(dsl_dataset_t *ds, nvlist_t *val) { uint64_t count = 0; objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset; zap_cursor_t zc; zap_attribute_t za; ASSERT(dsl_pool_config_held(ds->ds_dir->dd_pool)); /* * There may be missing entries in ds_next_clones_obj * due to a bug in a previous version of the code. * Only trust it if it has the right number of entries. */ if (dsl_dataset_phys(ds)->ds_next_clones_obj != 0) { VERIFY0(zap_count(mos, dsl_dataset_phys(ds)->ds_next_clones_obj, &count)); } if (count != dsl_dataset_phys(ds)->ds_num_children - 1) { return (SET_ERROR(ENOENT)); } for (zap_cursor_init(&zc, mos, dsl_dataset_phys(ds)->ds_next_clones_obj); zap_cursor_retrieve(&zc, &za) == 0; zap_cursor_advance(&zc)) { dsl_dataset_t *clone; char buf[ZFS_MAX_DATASET_NAME_LEN]; VERIFY0(dsl_dataset_hold_obj(ds->ds_dir->dd_pool, za.za_first_integer, FTAG, &clone)); dsl_dir_name(clone->ds_dir, buf); fnvlist_add_boolean(val, buf); dsl_dataset_rele(clone, FTAG); } zap_cursor_fini(&zc); return (0); } void get_clones_stat(dsl_dataset_t *ds, nvlist_t *nv) { nvlist_t *propval = fnvlist_alloc(); nvlist_t *val = fnvlist_alloc(); if (get_clones_stat_impl(ds, val) == 0) { fnvlist_add_nvlist(propval, ZPROP_VALUE, val); fnvlist_add_nvlist(nv, zfs_prop_to_name(ZFS_PROP_CLONES), propval); } nvlist_free(val); nvlist_free(propval); } static char * get_receive_resume_token_impl(dsl_dataset_t *ds) { if (!dsl_dataset_has_resume_receive_state(ds)) return (NULL); dsl_pool_t *dp = ds->ds_dir->dd_pool; char *str; void *packed; uint8_t *compressed; uint64_t val; nvlist_t *token_nv = fnvlist_alloc(); size_t packed_size, compressed_size; if (zap_lookup(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_FROMGUID, sizeof (val), 1, &val) == 0) { fnvlist_add_uint64(token_nv, "fromguid", val); } if (zap_lookup(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_OBJECT, sizeof (val), 1, &val) == 0) { fnvlist_add_uint64(token_nv, "object", val); } if (zap_lookup(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_OFFSET, sizeof (val), 1, &val) == 0) { fnvlist_add_uint64(token_nv, "offset", val); } if (zap_lookup(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_BYTES, sizeof (val), 1, &val) == 0) { fnvlist_add_uint64(token_nv, "bytes", val); } if (zap_lookup(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_TOGUID, sizeof (val), 1, &val) == 0) { fnvlist_add_uint64(token_nv, "toguid", val); } char buf[MAXNAMELEN]; if (zap_lookup(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_TONAME, 1, sizeof (buf), buf) == 0) { fnvlist_add_string(token_nv, "toname", buf); } if (zap_contains(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_LARGEBLOCK) == 0) { fnvlist_add_boolean(token_nv, "largeblockok"); } if (zap_contains(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_EMBEDOK) == 0) { fnvlist_add_boolean(token_nv, "embedok"); } if (zap_contains(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_COMPRESSOK) == 0) { fnvlist_add_boolean(token_nv, "compressok"); } if (zap_contains(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_RAWOK) == 0) { fnvlist_add_boolean(token_nv, "rawok"); } if (dsl_dataset_feature_is_active(ds, SPA_FEATURE_REDACTED_DATASETS)) { uint64_t num_redact_snaps = 0; uint64_t *redact_snaps = NULL; VERIFY3B(dsl_dataset_get_uint64_array_feature(ds, SPA_FEATURE_REDACTED_DATASETS, &num_redact_snaps, &redact_snaps), ==, B_TRUE); fnvlist_add_uint64_array(token_nv, "redact_snaps", redact_snaps, num_redact_snaps); } if (zap_contains(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_REDACT_BOOKMARK_SNAPS) == 0) { uint64_t num_redact_snaps = 0, int_size = 0; uint64_t *redact_snaps = NULL; VERIFY0(zap_length(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_REDACT_BOOKMARK_SNAPS, &int_size, &num_redact_snaps)); ASSERT3U(int_size, ==, sizeof (uint64_t)); redact_snaps = kmem_alloc(int_size * num_redact_snaps, KM_SLEEP); VERIFY0(zap_lookup(dp->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_REDACT_BOOKMARK_SNAPS, int_size, num_redact_snaps, redact_snaps)); fnvlist_add_uint64_array(token_nv, "book_redact_snaps", redact_snaps, num_redact_snaps); kmem_free(redact_snaps, int_size * num_redact_snaps); } packed = fnvlist_pack(token_nv, &packed_size); fnvlist_free(token_nv); compressed = kmem_alloc(packed_size, KM_SLEEP); /* Call compress function directly to avoid hole detection. */ - compressed_size = zfs_gzip_compress(packed, compressed, + abd_t pabd, cabd; + abd_get_from_buf_struct(&pabd, packed, packed_size); + abd_get_from_buf_struct(&cabd, compressed, packed_size); + compressed_size = zfs_gzip_compress(&pabd, &cabd, packed_size, packed_size, 6); + abd_free(&cabd); + abd_free(&pabd); zio_cksum_t cksum; fletcher_4_native_varsize(compressed, compressed_size, &cksum); size_t alloc_size = compressed_size * 2 + 1; str = kmem_alloc(alloc_size, KM_SLEEP); for (int i = 0; i < compressed_size; i++) { size_t offset = i * 2; (void) snprintf(str + offset, alloc_size - offset, "%02x", compressed[i]); } str[compressed_size * 2] = '\0'; char *propval = kmem_asprintf("%u-%llx-%llx-%s", ZFS_SEND_RESUME_TOKEN_VERSION, (longlong_t)cksum.zc_word[0], (longlong_t)packed_size, str); kmem_free(packed, packed_size); kmem_free(str, alloc_size); kmem_free(compressed, packed_size); return (propval); } /* * Returns a string that represents the receive resume state token. It should * be freed with strfree(). NULL is returned if no resume state is present. */ char * get_receive_resume_token(dsl_dataset_t *ds) { /* * A failed "newfs" (e.g. full) resumable receive leaves * the stats set on this dataset. Check here for the prop. */ char *token = get_receive_resume_token_impl(ds); if (token != NULL) return (token); /* * A failed incremental resumable receive leaves the * stats set on our child named "%recv". Check the child * for the prop. */ /* 6 extra bytes for /%recv */ char name[ZFS_MAX_DATASET_NAME_LEN + 6]; dsl_dataset_t *recv_ds; dsl_dataset_name(ds, name); if (strlcat(name, "/", sizeof (name)) < sizeof (name) && strlcat(name, recv_clone_name, sizeof (name)) < sizeof (name) && dsl_dataset_hold(ds->ds_dir->dd_pool, name, FTAG, &recv_ds) == 0) { token = get_receive_resume_token_impl(recv_ds); dsl_dataset_rele(recv_ds, FTAG); } return (token); } uint64_t dsl_get_refratio(dsl_dataset_t *ds) { uint64_t ratio = dsl_dataset_phys(ds)->ds_compressed_bytes == 0 ? 100 : (dsl_dataset_phys(ds)->ds_uncompressed_bytes * 100 / dsl_dataset_phys(ds)->ds_compressed_bytes); return (ratio); } uint64_t dsl_get_logicalreferenced(dsl_dataset_t *ds) { return (dsl_dataset_phys(ds)->ds_uncompressed_bytes); } uint64_t dsl_get_compressratio(dsl_dataset_t *ds) { if (ds->ds_is_snapshot) { return (dsl_get_refratio(ds)); } else { dsl_dir_t *dd = ds->ds_dir; mutex_enter(&dd->dd_lock); uint64_t val = dsl_dir_get_compressratio(dd); mutex_exit(&dd->dd_lock); return (val); } } uint64_t dsl_get_used(dsl_dataset_t *ds) { if (ds->ds_is_snapshot) { return (dsl_dataset_phys(ds)->ds_unique_bytes); } else { dsl_dir_t *dd = ds->ds_dir; mutex_enter(&dd->dd_lock); uint64_t val = dsl_dir_get_used(dd); mutex_exit(&dd->dd_lock); return (val); } } uint64_t dsl_get_creation(dsl_dataset_t *ds) { return (dsl_dataset_phys(ds)->ds_creation_time); } uint64_t dsl_get_creationtxg(dsl_dataset_t *ds) { return (dsl_dataset_phys(ds)->ds_creation_txg); } uint64_t dsl_get_refquota(dsl_dataset_t *ds) { return (ds->ds_quota); } uint64_t dsl_get_refreservation(dsl_dataset_t *ds) { return (ds->ds_reserved); } uint64_t dsl_get_guid(dsl_dataset_t *ds) { return (dsl_dataset_phys(ds)->ds_guid); } uint64_t dsl_get_unique(dsl_dataset_t *ds) { return (dsl_dataset_phys(ds)->ds_unique_bytes); } uint64_t dsl_get_objsetid(dsl_dataset_t *ds) { return (ds->ds_object); } uint64_t dsl_get_userrefs(dsl_dataset_t *ds) { return (ds->ds_userrefs); } uint64_t dsl_get_defer_destroy(dsl_dataset_t *ds) { return (DS_IS_DEFER_DESTROY(ds) ? 1 : 0); } uint64_t dsl_get_referenced(dsl_dataset_t *ds) { return (dsl_dataset_phys(ds)->ds_referenced_bytes); } uint64_t dsl_get_numclones(dsl_dataset_t *ds) { ASSERT(ds->ds_is_snapshot); return (dsl_dataset_phys(ds)->ds_num_children - 1); } uint64_t dsl_get_inconsistent(dsl_dataset_t *ds) { return ((dsl_dataset_phys(ds)->ds_flags & DS_FLAG_INCONSISTENT) ? 1 : 0); } uint64_t dsl_get_redacted(dsl_dataset_t *ds) { return (dsl_dataset_feature_is_active(ds, SPA_FEATURE_REDACTED_DATASETS)); } uint64_t dsl_get_available(dsl_dataset_t *ds) { uint64_t refdbytes = dsl_get_referenced(ds); uint64_t availbytes = dsl_dir_space_available(ds->ds_dir, NULL, 0, TRUE); if (ds->ds_reserved > dsl_dataset_phys(ds)->ds_unique_bytes) { availbytes += ds->ds_reserved - dsl_dataset_phys(ds)->ds_unique_bytes; } if (ds->ds_quota != 0) { /* * Adjust available bytes according to refquota */ if (refdbytes < ds->ds_quota) { availbytes = MIN(availbytes, ds->ds_quota - refdbytes); } else { availbytes = 0; } } return (availbytes); } int dsl_get_written(dsl_dataset_t *ds, uint64_t *written) { dsl_pool_t *dp = ds->ds_dir->dd_pool; dsl_dataset_t *prev; int err = dsl_dataset_hold_obj(dp, dsl_dataset_phys(ds)->ds_prev_snap_obj, FTAG, &prev); if (err == 0) { uint64_t comp, uncomp; err = dsl_dataset_space_written(prev, ds, written, &comp, &uncomp); dsl_dataset_rele(prev, FTAG); } return (err); } /* * 'snap' should be a buffer of size ZFS_MAX_DATASET_NAME_LEN. */ int dsl_get_prev_snap(dsl_dataset_t *ds, char *snap) { dsl_pool_t *dp = ds->ds_dir->dd_pool; if (ds->ds_prev != NULL && ds->ds_prev != dp->dp_origin_snap) { dsl_dataset_name(ds->ds_prev, snap); return (0); } else { return (SET_ERROR(ENOENT)); } } void dsl_get_redact_snaps(dsl_dataset_t *ds, nvlist_t *propval) { uint64_t nsnaps; uint64_t *snaps; if (dsl_dataset_get_uint64_array_feature(ds, SPA_FEATURE_REDACTED_DATASETS, &nsnaps, &snaps)) { fnvlist_add_uint64_array(propval, ZPROP_VALUE, snaps, nsnaps); } } /* * Returns the mountpoint property and source for the given dataset in the value * and source buffers. The value buffer must be at least as large as MAXPATHLEN * and the source buffer as least as large a ZFS_MAX_DATASET_NAME_LEN. * Returns 0 on success and an error on failure. */ int dsl_get_mountpoint(dsl_dataset_t *ds, const char *dsname, char *value, char *source) { int error; dsl_pool_t *dp = ds->ds_dir->dd_pool; /* Retrieve the mountpoint value stored in the zap object */ error = dsl_prop_get_ds(ds, zfs_prop_to_name(ZFS_PROP_MOUNTPOINT), 1, ZAP_MAXVALUELEN, value, source); if (error != 0) { return (error); } /* * Process the dsname and source to find the full mountpoint string. * Can be skipped for 'legacy' or 'none'. */ if (value[0] == '/') { char *buf = kmem_alloc(ZAP_MAXVALUELEN, KM_SLEEP); char *root = buf; const char *relpath; /* * If we inherit the mountpoint, even from a dataset * with a received value, the source will be the path of * the dataset we inherit from. If source is * ZPROP_SOURCE_VAL_RECVD, the received value is not * inherited. */ if (strcmp(source, ZPROP_SOURCE_VAL_RECVD) == 0) { relpath = ""; } else { ASSERT0(strncmp(dsname, source, strlen(source))); relpath = dsname + strlen(source); if (relpath[0] == '/') relpath++; } spa_altroot(dp->dp_spa, root, ZAP_MAXVALUELEN); /* * Special case an alternate root of '/'. This will * avoid having multiple leading slashes in the * mountpoint path. */ if (strcmp(root, "/") == 0) root++; /* * If the mountpoint is '/' then skip over this * if we are obtaining either an alternate root or * an inherited mountpoint. */ char *mnt = value; if (value[1] == '\0' && (root[0] != '\0' || relpath[0] != '\0')) mnt = value + 1; mnt = kmem_strdup(mnt); if (relpath[0] == '\0') { (void) snprintf(value, ZAP_MAXVALUELEN, "%s%s", root, mnt); } else { (void) snprintf(value, ZAP_MAXVALUELEN, "%s%s%s%s", root, mnt, relpath[0] == '@' ? "" : "/", relpath); } kmem_free(buf, ZAP_MAXVALUELEN); kmem_strfree(mnt); } return (0); } void dsl_dataset_stats(dsl_dataset_t *ds, nvlist_t *nv) { dsl_pool_t *dp __maybe_unused = ds->ds_dir->dd_pool; ASSERT(dsl_pool_config_held(dp)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFRATIO, dsl_get_refratio(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_LOGICALREFERENCED, dsl_get_logicalreferenced(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_COMPRESSRATIO, dsl_get_compressratio(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_USED, dsl_get_used(ds)); if (ds->ds_is_snapshot) { get_clones_stat(ds, nv); } else { char buf[ZFS_MAX_DATASET_NAME_LEN]; if (dsl_get_prev_snap(ds, buf) == 0) dsl_prop_nvlist_add_string(nv, ZFS_PROP_PREV_SNAP, buf); dsl_dir_stats(ds->ds_dir, nv); } nvlist_t *propval = fnvlist_alloc(); dsl_get_redact_snaps(ds, propval); fnvlist_add_nvlist(nv, zfs_prop_to_name(ZFS_PROP_REDACT_SNAPS), propval); nvlist_free(propval); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_AVAILABLE, dsl_get_available(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFERENCED, dsl_get_referenced(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_CREATION, dsl_get_creation(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_CREATETXG, dsl_get_creationtxg(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFQUOTA, dsl_get_refquota(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_REFRESERVATION, dsl_get_refreservation(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_GUID, dsl_get_guid(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_UNIQUE, dsl_get_unique(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_OBJSETID, dsl_get_objsetid(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_USERREFS, dsl_get_userrefs(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_DEFER_DESTROY, dsl_get_defer_destroy(ds)); dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_SNAPSHOTS_CHANGED, dsl_dir_snap_cmtime(ds->ds_dir).tv_sec); dsl_dataset_crypt_stats(ds, nv); if (dsl_dataset_phys(ds)->ds_prev_snap_obj != 0) { uint64_t written; if (dsl_get_written(ds, &written) == 0) { dsl_prop_nvlist_add_uint64(nv, ZFS_PROP_WRITTEN, written); } } if (!dsl_dataset_is_snapshot(ds)) { char *token = get_receive_resume_token(ds); if (token != NULL) { dsl_prop_nvlist_add_string(nv, ZFS_PROP_RECEIVE_RESUME_TOKEN, token); kmem_strfree(token); } } } void dsl_dataset_fast_stat(dsl_dataset_t *ds, dmu_objset_stats_t *stat) { dsl_pool_t *dp __maybe_unused = ds->ds_dir->dd_pool; ASSERT(dsl_pool_config_held(dp)); stat->dds_creation_txg = dsl_get_creationtxg(ds); stat->dds_inconsistent = dsl_get_inconsistent(ds); stat->dds_guid = dsl_get_guid(ds); stat->dds_redacted = dsl_get_redacted(ds); stat->dds_origin[0] = '\0'; if (ds->ds_is_snapshot) { stat->dds_is_snapshot = B_TRUE; stat->dds_num_clones = dsl_get_numclones(ds); } else { stat->dds_is_snapshot = B_FALSE; stat->dds_num_clones = 0; if (dsl_dir_is_clone(ds->ds_dir)) { dsl_dir_get_origin(ds->ds_dir, stat->dds_origin); } } } uint64_t dsl_dataset_fsid_guid(dsl_dataset_t *ds) { return (ds->ds_fsid_guid); } void dsl_dataset_space(dsl_dataset_t *ds, uint64_t *refdbytesp, uint64_t *availbytesp, uint64_t *usedobjsp, uint64_t *availobjsp) { *refdbytesp = dsl_dataset_phys(ds)->ds_referenced_bytes; *availbytesp = dsl_dir_space_available(ds->ds_dir, NULL, 0, TRUE); if (ds->ds_reserved > dsl_dataset_phys(ds)->ds_unique_bytes) *availbytesp += ds->ds_reserved - dsl_dataset_phys(ds)->ds_unique_bytes; if (ds->ds_quota != 0) { /* * Adjust available bytes according to refquota */ if (*refdbytesp < ds->ds_quota) *availbytesp = MIN(*availbytesp, ds->ds_quota - *refdbytesp); else *availbytesp = 0; } rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG); *usedobjsp = BP_GET_FILL(&dsl_dataset_phys(ds)->ds_bp); rrw_exit(&ds->ds_bp_rwlock, FTAG); *availobjsp = DN_MAX_OBJECT - *usedobjsp; } boolean_t dsl_dataset_modified_since_snap(dsl_dataset_t *ds, dsl_dataset_t *snap) { dsl_pool_t *dp __maybe_unused = ds->ds_dir->dd_pool; uint64_t birth; ASSERT(dsl_pool_config_held(dp)); if (snap == NULL) return (B_FALSE); rrw_enter(&ds->ds_bp_rwlock, RW_READER, FTAG); birth = BP_GET_LOGICAL_BIRTH(dsl_dataset_get_blkptr(ds)); rrw_exit(&ds->ds_bp_rwlock, FTAG); if (birth > dsl_dataset_phys(snap)->ds_creation_txg) { objset_t *os, *os_snap; /* * It may be that only the ZIL differs, because it was * reset in the head. Don't count that as being * modified. */ if (dmu_objset_from_ds(ds, &os) != 0) return (B_TRUE); if (dmu_objset_from_ds(snap, &os_snap) != 0) return (B_TRUE); return (memcmp(&os->os_phys->os_meta_dnode, &os_snap->os_phys->os_meta_dnode, sizeof (os->os_phys->os_meta_dnode)) != 0); } return (B_FALSE); } static int dsl_dataset_rename_snapshot_check_impl(dsl_pool_t *dp, dsl_dataset_t *hds, void *arg) { (void) dp; dsl_dataset_rename_snapshot_arg_t *ddrsa = arg; int error; uint64_t val; error = dsl_dataset_snap_lookup(hds, ddrsa->ddrsa_oldsnapname, &val); if (error != 0) { /* ignore nonexistent snapshots */ return (error == ENOENT ? 0 : error); } /* new name should not exist */ error = dsl_dataset_snap_lookup(hds, ddrsa->ddrsa_newsnapname, &val); if (error == 0) error = SET_ERROR(EEXIST); else if (error == ENOENT) error = 0; /* dataset name + 1 for the "@" + the new snapshot name must fit */ if (dsl_dir_namelen(hds->ds_dir) + 1 + strlen(ddrsa->ddrsa_newsnapname) >= ZFS_MAX_DATASET_NAME_LEN) error = SET_ERROR(ENAMETOOLONG); return (error); } int dsl_dataset_rename_snapshot_check(void *arg, dmu_tx_t *tx) { dsl_dataset_rename_snapshot_arg_t *ddrsa = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *hds; int error; error = dsl_dataset_hold(dp, ddrsa->ddrsa_fsname, FTAG, &hds); if (error != 0) return (error); if (ddrsa->ddrsa_recursive) { error = dmu_objset_find_dp(dp, hds->ds_dir->dd_object, dsl_dataset_rename_snapshot_check_impl, ddrsa, DS_FIND_CHILDREN); } else { error = dsl_dataset_rename_snapshot_check_impl(dp, hds, ddrsa); } dsl_dataset_rele(hds, FTAG); return (error); } static int dsl_dataset_rename_snapshot_sync_impl(dsl_pool_t *dp, dsl_dataset_t *hds, void *arg) { dsl_dataset_rename_snapshot_arg_t *ddrsa = arg; dsl_dataset_t *ds; uint64_t val; dmu_tx_t *tx = ddrsa->ddrsa_tx; int error; error = dsl_dataset_snap_lookup(hds, ddrsa->ddrsa_oldsnapname, &val); ASSERT(error == 0 || error == ENOENT); if (error == ENOENT) { /* ignore nonexistent snapshots */ return (0); } VERIFY0(dsl_dataset_hold_obj(dp, val, FTAG, &ds)); /* log before we change the name */ spa_history_log_internal_ds(ds, "rename", tx, "-> @%s", ddrsa->ddrsa_newsnapname); VERIFY0(dsl_dataset_snap_remove(hds, ddrsa->ddrsa_oldsnapname, tx, B_FALSE)); mutex_enter(&ds->ds_lock); (void) strlcpy(ds->ds_snapname, ddrsa->ddrsa_newsnapname, sizeof (ds->ds_snapname)); mutex_exit(&ds->ds_lock); VERIFY0(zap_add(dp->dp_meta_objset, dsl_dataset_phys(hds)->ds_snapnames_zapobj, ds->ds_snapname, 8, 1, &ds->ds_object, tx)); zvol_rename_minors(dp->dp_spa, ddrsa->ddrsa_oldsnapname, ddrsa->ddrsa_newsnapname, B_TRUE); dsl_dataset_rele(ds, FTAG); return (0); } void dsl_dataset_rename_snapshot_sync(void *arg, dmu_tx_t *tx) { dsl_dataset_rename_snapshot_arg_t *ddrsa = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *hds = NULL; VERIFY0(dsl_dataset_hold(dp, ddrsa->ddrsa_fsname, FTAG, &hds)); ddrsa->ddrsa_tx = tx; if (ddrsa->ddrsa_recursive) { VERIFY0(dmu_objset_find_dp(dp, hds->ds_dir->dd_object, dsl_dataset_rename_snapshot_sync_impl, ddrsa, DS_FIND_CHILDREN)); } else { VERIFY0(dsl_dataset_rename_snapshot_sync_impl(dp, hds, ddrsa)); } dsl_dataset_rele(hds, FTAG); } int dsl_dataset_rename_snapshot(const char *fsname, const char *oldsnapname, const char *newsnapname, boolean_t recursive) { dsl_dataset_rename_snapshot_arg_t ddrsa; ddrsa.ddrsa_fsname = fsname; ddrsa.ddrsa_oldsnapname = oldsnapname; ddrsa.ddrsa_newsnapname = newsnapname; ddrsa.ddrsa_recursive = recursive; return (dsl_sync_task(fsname, dsl_dataset_rename_snapshot_check, dsl_dataset_rename_snapshot_sync, &ddrsa, 1, ZFS_SPACE_CHECK_RESERVED)); } /* * If we're doing an ownership handoff, we need to make sure that there is * only one long hold on the dataset. We're not allowed to change anything here * so we don't permanently release the long hold or regular hold here. We want * to do this only when syncing to avoid the dataset unexpectedly going away * when we release the long hold. */ static int dsl_dataset_handoff_check(dsl_dataset_t *ds, void *owner, dmu_tx_t *tx) { boolean_t held = B_FALSE; if (!dmu_tx_is_syncing(tx)) return (0); dsl_dir_t *dd = ds->ds_dir; mutex_enter(&dd->dd_activity_lock); uint64_t holds = zfs_refcount_count(&ds->ds_longholds) - (owner != NULL ? 1 : 0); /* * The value of dd_activity_waiters can chance as soon as we drop the * lock, but we're fine with that; new waiters coming in or old * waiters leaving doesn't cause problems, since we're going to cancel * waiters later anyway. The goal of this check is to verify that no * non-waiters have long-holds, and all new long-holds will be * prevented because we're holding the pool config as writer. */ if (holds != dd->dd_activity_waiters) held = B_TRUE; mutex_exit(&dd->dd_activity_lock); if (held) return (SET_ERROR(EBUSY)); return (0); } int dsl_dataset_rollback_check(void *arg, dmu_tx_t *tx) { dsl_dataset_rollback_arg_t *ddra = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *ds; int64_t unused_refres_delta; int error; error = dsl_dataset_hold(dp, ddra->ddra_fsname, FTAG, &ds); if (error != 0) return (error); /* must not be a snapshot */ if (ds->ds_is_snapshot) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(EINVAL)); } /* must have a most recent snapshot */ if (dsl_dataset_phys(ds)->ds_prev_snap_txg < TXG_INITIAL) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(ESRCH)); } /* * No rollback to a snapshot created in the current txg, because * the rollback may dirty the dataset and create blocks that are * not reachable from the rootbp while having a birth txg that * falls into the snapshot's range. */ if (dmu_tx_is_syncing(tx) && dsl_dataset_phys(ds)->ds_prev_snap_txg >= tx->tx_txg) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(EAGAIN)); } /* * If the expected target snapshot is specified, then check that * the latest snapshot is it. */ if (ddra->ddra_tosnap != NULL) { dsl_dataset_t *snapds; /* Check if the target snapshot exists at all. */ error = dsl_dataset_hold(dp, ddra->ddra_tosnap, FTAG, &snapds); if (error != 0) { /* * ESRCH is used to signal that the target snapshot does * not exist, while ENOENT is used to report that * the rolled back dataset does not exist. * ESRCH is also used to cover other cases where the * target snapshot is not related to the dataset being * rolled back such as being in a different pool. */ if (error == ENOENT || error == EXDEV) error = SET_ERROR(ESRCH); dsl_dataset_rele(ds, FTAG); return (error); } ASSERT(snapds->ds_is_snapshot); /* Check if the snapshot is the latest snapshot indeed. */ if (snapds != ds->ds_prev) { /* * Distinguish between the case where the only problem * is intervening snapshots (EEXIST) vs the snapshot * not being a valid target for rollback (ESRCH). */ if (snapds->ds_dir == ds->ds_dir || (dsl_dir_is_clone(ds->ds_dir) && dsl_dir_phys(ds->ds_dir)->dd_origin_obj == snapds->ds_object)) { error = SET_ERROR(EEXIST); } else { error = SET_ERROR(ESRCH); } dsl_dataset_rele(snapds, FTAG); dsl_dataset_rele(ds, FTAG); return (error); } dsl_dataset_rele(snapds, FTAG); } /* must not have any bookmarks after the most recent snapshot */ if (dsl_bookmark_latest_txg(ds) > dsl_dataset_phys(ds)->ds_prev_snap_txg) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(EEXIST)); } error = dsl_dataset_handoff_check(ds, ddra->ddra_owner, tx); if (error != 0) { dsl_dataset_rele(ds, FTAG); return (error); } /* * Check if the snap we are rolling back to uses more than * the refquota. */ if (ds->ds_quota != 0 && dsl_dataset_phys(ds->ds_prev)->ds_referenced_bytes > ds->ds_quota) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(EDQUOT)); } /* * When we do the clone swap, we will temporarily use more space * due to the refreservation (the head will no longer have any * unique space, so the entire amount of the refreservation will need * to be free). We will immediately destroy the clone, freeing * this space, but the freeing happens over many txg's. */ unused_refres_delta = (int64_t)MIN(ds->ds_reserved, dsl_dataset_phys(ds)->ds_unique_bytes); if (unused_refres_delta > 0 && unused_refres_delta > dsl_dir_space_available(ds->ds_dir, NULL, 0, TRUE)) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(ENOSPC)); } dsl_dataset_rele(ds, FTAG); return (0); } void dsl_dataset_rollback_sync(void *arg, dmu_tx_t *tx) { dsl_dataset_rollback_arg_t *ddra = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *ds, *clone; uint64_t cloneobj; char namebuf[ZFS_MAX_DATASET_NAME_LEN]; VERIFY0(dsl_dataset_hold(dp, ddra->ddra_fsname, FTAG, &ds)); dsl_dataset_name(ds->ds_prev, namebuf); fnvlist_add_string(ddra->ddra_result, "target", namebuf); cloneobj = dsl_dataset_create_sync(ds->ds_dir, "%rollback", ds->ds_prev, DS_CREATE_FLAG_NODIRTY, kcred, NULL, tx); VERIFY0(dsl_dataset_hold_obj(dp, cloneobj, FTAG, &clone)); dsl_dataset_clone_swap_sync_impl(clone, ds, tx); dsl_dataset_zero_zil(ds, tx); dsl_destroy_head_sync_impl(clone, tx); dsl_dataset_rele(clone, FTAG); dsl_dataset_rele(ds, FTAG); } /* * Rolls back the given filesystem or volume to the most recent snapshot. * The name of the most recent snapshot will be returned under key "target" * in the result nvlist. * * If owner != NULL: * - The existing dataset MUST be owned by the specified owner at entry * - Upon return, dataset will still be held by the same owner, whether we * succeed or not. * * This mode is required any time the existing filesystem is mounted. See * notes above zfs_suspend_fs() for further details. */ int dsl_dataset_rollback(const char *fsname, const char *tosnap, void *owner, nvlist_t *result) { dsl_dataset_rollback_arg_t ddra; ddra.ddra_fsname = fsname; ddra.ddra_tosnap = tosnap; ddra.ddra_owner = owner; ddra.ddra_result = result; return (dsl_sync_task(fsname, dsl_dataset_rollback_check, dsl_dataset_rollback_sync, &ddra, 1, ZFS_SPACE_CHECK_RESERVED)); } struct promotenode { list_node_t link; dsl_dataset_t *ds; }; static int snaplist_space(list_t *l, uint64_t mintxg, uint64_t *spacep); static int promote_hold(dsl_dataset_promote_arg_t *ddpa, dsl_pool_t *dp, const void *tag); static void promote_rele(dsl_dataset_promote_arg_t *ddpa, const void *tag); int dsl_dataset_promote_check(void *arg, dmu_tx_t *tx) { dsl_dataset_promote_arg_t *ddpa = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *hds; struct promotenode *snap; int err; uint64_t unused; uint64_t ss_mv_cnt; size_t max_snap_len; boolean_t conflicting_snaps; err = promote_hold(ddpa, dp, FTAG); if (err != 0) return (err); hds = ddpa->ddpa_clone; max_snap_len = MAXNAMELEN - strlen(ddpa->ddpa_clonename) - 1; if (dsl_dataset_phys(hds)->ds_flags & DS_FLAG_NOPROMOTE) { promote_rele(ddpa, FTAG); return (SET_ERROR(EXDEV)); } snap = list_head(&ddpa->shared_snaps); if (snap == NULL) { err = SET_ERROR(ENOENT); goto out; } dsl_dataset_t *const origin_ds = snap->ds; /* * Encrypted clones share a DSL Crypto Key with their origin's dsl dir. * When doing a promote we must make sure the encryption root for * both the target and the target's origin does not change to avoid * needing to rewrap encryption keys */ err = dsl_dataset_promote_crypt_check(hds->ds_dir, origin_ds->ds_dir); if (err != 0) goto out; /* * Compute and check the amount of space to transfer. Since this is * so expensive, don't do the preliminary check. */ if (!dmu_tx_is_syncing(tx)) { promote_rele(ddpa, FTAG); return (0); } /* compute origin's new unique space */ snap = list_tail(&ddpa->clone_snaps); ASSERT(snap != NULL); ASSERT3U(dsl_dataset_phys(snap->ds)->ds_prev_snap_obj, ==, origin_ds->ds_object); dsl_deadlist_space_range(&snap->ds->ds_deadlist, dsl_dataset_phys(origin_ds)->ds_prev_snap_txg, UINT64_MAX, &ddpa->unique, &unused, &unused); /* * Walk the snapshots that we are moving * * Compute space to transfer. Consider the incremental changes * to used by each snapshot: * (my used) = (prev's used) + (blocks born) - (blocks killed) * So each snapshot gave birth to: * (blocks born) = (my used) - (prev's used) + (blocks killed) * So a sequence would look like: * (uN - u(N-1) + kN) + ... + (u1 - u0 + k1) + (u0 - 0 + k0) * Which simplifies to: * uN + kN + kN-1 + ... + k1 + k0 * Note however, if we stop before we reach the ORIGIN we get: * uN + kN + kN-1 + ... + kM - uM-1 */ conflicting_snaps = B_FALSE; ss_mv_cnt = 0; ddpa->used = dsl_dataset_phys(origin_ds)->ds_referenced_bytes; ddpa->comp = dsl_dataset_phys(origin_ds)->ds_compressed_bytes; ddpa->uncomp = dsl_dataset_phys(origin_ds)->ds_uncompressed_bytes; for (snap = list_head(&ddpa->shared_snaps); snap; snap = list_next(&ddpa->shared_snaps, snap)) { uint64_t val, dlused, dlcomp, dluncomp; dsl_dataset_t *ds = snap->ds; ss_mv_cnt++; /* * If there are long holds, we won't be able to evict * the objset. */ if (dsl_dataset_long_held(ds)) { err = SET_ERROR(EBUSY); goto out; } /* Check that the snapshot name does not conflict */ VERIFY0(dsl_dataset_get_snapname(ds)); if (strlen(ds->ds_snapname) >= max_snap_len) { err = SET_ERROR(ENAMETOOLONG); goto out; } err = dsl_dataset_snap_lookup(hds, ds->ds_snapname, &val); if (err == 0) { fnvlist_add_boolean(ddpa->err_ds, snap->ds->ds_snapname); conflicting_snaps = B_TRUE; } else if (err != ENOENT) { goto out; } /* The very first snapshot does not have a deadlist */ if (dsl_dataset_phys(ds)->ds_prev_snap_obj == 0) continue; dsl_deadlist_space(&ds->ds_deadlist, &dlused, &dlcomp, &dluncomp); ddpa->used += dlused; ddpa->comp += dlcomp; ddpa->uncomp += dluncomp; } /* * Check that bookmarks that are being transferred don't have * name conflicts. */ for (dsl_bookmark_node_t *dbn = avl_first(&origin_ds->ds_bookmarks); dbn != NULL && dbn->dbn_phys.zbm_creation_txg <= dsl_dataset_phys(origin_ds)->ds_creation_txg; dbn = AVL_NEXT(&origin_ds->ds_bookmarks, dbn)) { if (strlen(dbn->dbn_name) >= max_snap_len) { err = SET_ERROR(ENAMETOOLONG); goto out; } zfs_bookmark_phys_t bm; err = dsl_bookmark_lookup_impl(ddpa->ddpa_clone, dbn->dbn_name, &bm); if (err == 0) { fnvlist_add_boolean(ddpa->err_ds, dbn->dbn_name); conflicting_snaps = B_TRUE; } else if (err == ESRCH) { err = 0; } if (err != 0) { goto out; } } /* * In order to return the full list of conflicting snapshots, we check * whether there was a conflict after traversing all of them. */ if (conflicting_snaps) { err = SET_ERROR(EEXIST); goto out; } /* * If we are a clone of a clone then we never reached ORIGIN, * so we need to subtract out the clone origin's used space. */ if (ddpa->origin_origin) { ddpa->used -= dsl_dataset_phys(ddpa->origin_origin)->ds_referenced_bytes; ddpa->comp -= dsl_dataset_phys(ddpa->origin_origin)->ds_compressed_bytes; ddpa->uncomp -= dsl_dataset_phys(ddpa->origin_origin)-> ds_uncompressed_bytes; } /* Check that there is enough space and limit headroom here */ err = dsl_dir_transfer_possible(origin_ds->ds_dir, hds->ds_dir, 0, ss_mv_cnt, ddpa->used, ddpa->cr, ddpa->proc); if (err != 0) goto out; /* * Compute the amounts of space that will be used by snapshots * after the promotion (for both origin and clone). For each, * it is the amount of space that will be on all of their * deadlists (that was not born before their new origin). */ if (dsl_dir_phys(hds->ds_dir)->dd_flags & DD_FLAG_USED_BREAKDOWN) { uint64_t space; /* * Note, typically this will not be a clone of a clone, * so dd_origin_txg will be < TXG_INITIAL, so * these snaplist_space() -> dsl_deadlist_space_range() * calls will be fast because they do not have to * iterate over all bps. */ snap = list_head(&ddpa->origin_snaps); if (snap == NULL) { err = SET_ERROR(ENOENT); goto out; } err = snaplist_space(&ddpa->shared_snaps, snap->ds->ds_dir->dd_origin_txg, &ddpa->cloneusedsnap); if (err != 0) goto out; err = snaplist_space(&ddpa->clone_snaps, snap->ds->ds_dir->dd_origin_txg, &space); if (err != 0) goto out; ddpa->cloneusedsnap += space; } if (dsl_dir_phys(origin_ds->ds_dir)->dd_flags & DD_FLAG_USED_BREAKDOWN) { err = snaplist_space(&ddpa->origin_snaps, dsl_dataset_phys(origin_ds)->ds_creation_txg, &ddpa->originusedsnap); if (err != 0) goto out; } out: promote_rele(ddpa, FTAG); return (err); } void dsl_dataset_promote_sync(void *arg, dmu_tx_t *tx) { dsl_dataset_promote_arg_t *ddpa = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *hds; struct promotenode *snap; dsl_dataset_t *origin_ds; dsl_dataset_t *origin_head; dsl_dir_t *dd; dsl_dir_t *odd = NULL; uint64_t oldnext_obj; int64_t delta; ASSERT(nvlist_empty(ddpa->err_ds)); VERIFY0(promote_hold(ddpa, dp, FTAG)); hds = ddpa->ddpa_clone; ASSERT0(dsl_dataset_phys(hds)->ds_flags & DS_FLAG_NOPROMOTE); snap = list_head(&ddpa->shared_snaps); origin_ds = snap->ds; dd = hds->ds_dir; snap = list_head(&ddpa->origin_snaps); origin_head = snap->ds; /* * We need to explicitly open odd, since origin_ds's dd will be * changing. */ VERIFY0(dsl_dir_hold_obj(dp, origin_ds->ds_dir->dd_object, NULL, FTAG, &odd)); dsl_dataset_promote_crypt_sync(hds->ds_dir, odd, tx); /* change origin's next snap */ dmu_buf_will_dirty(origin_ds->ds_dbuf, tx); oldnext_obj = dsl_dataset_phys(origin_ds)->ds_next_snap_obj; snap = list_tail(&ddpa->clone_snaps); ASSERT3U(dsl_dataset_phys(snap->ds)->ds_prev_snap_obj, ==, origin_ds->ds_object); dsl_dataset_phys(origin_ds)->ds_next_snap_obj = snap->ds->ds_object; /* change the origin's next clone */ if (dsl_dataset_phys(origin_ds)->ds_next_clones_obj) { dsl_dataset_remove_from_next_clones(origin_ds, snap->ds->ds_object, tx); VERIFY0(zap_add_int(dp->dp_meta_objset, dsl_dataset_phys(origin_ds)->ds_next_clones_obj, oldnext_obj, tx)); } /* change origin */ dmu_buf_will_dirty(dd->dd_dbuf, tx); ASSERT3U(dsl_dir_phys(dd)->dd_origin_obj, ==, origin_ds->ds_object); dsl_dir_phys(dd)->dd_origin_obj = dsl_dir_phys(odd)->dd_origin_obj; dd->dd_origin_txg = origin_head->ds_dir->dd_origin_txg; dmu_buf_will_dirty(odd->dd_dbuf, tx); dsl_dir_phys(odd)->dd_origin_obj = origin_ds->ds_object; origin_head->ds_dir->dd_origin_txg = dsl_dataset_phys(origin_ds)->ds_creation_txg; /* change dd_clone entries */ if (spa_version(dp->dp_spa) >= SPA_VERSION_DIR_CLONES) { VERIFY0(zap_remove_int(dp->dp_meta_objset, dsl_dir_phys(odd)->dd_clones, hds->ds_object, tx)); VERIFY0(zap_add_int(dp->dp_meta_objset, dsl_dir_phys(ddpa->origin_origin->ds_dir)->dd_clones, hds->ds_object, tx)); VERIFY0(zap_remove_int(dp->dp_meta_objset, dsl_dir_phys(ddpa->origin_origin->ds_dir)->dd_clones, origin_head->ds_object, tx)); if (dsl_dir_phys(dd)->dd_clones == 0) { dsl_dir_phys(dd)->dd_clones = zap_create(dp->dp_meta_objset, DMU_OT_DSL_CLONES, DMU_OT_NONE, 0, tx); } VERIFY0(zap_add_int(dp->dp_meta_objset, dsl_dir_phys(dd)->dd_clones, origin_head->ds_object, tx)); } /* * Move bookmarks to this dir. */ dsl_bookmark_node_t *dbn_next; for (dsl_bookmark_node_t *dbn = avl_first(&origin_head->ds_bookmarks); dbn != NULL && dbn->dbn_phys.zbm_creation_txg <= dsl_dataset_phys(origin_ds)->ds_creation_txg; dbn = dbn_next) { dbn_next = AVL_NEXT(&origin_head->ds_bookmarks, dbn); avl_remove(&origin_head->ds_bookmarks, dbn); VERIFY0(zap_remove(dp->dp_meta_objset, origin_head->ds_bookmarks_obj, dbn->dbn_name, tx)); dsl_bookmark_node_add(hds, dbn, tx); } dsl_bookmark_next_changed(hds, origin_ds, tx); /* move snapshots to this dir */ for (snap = list_head(&ddpa->shared_snaps); snap; snap = list_next(&ddpa->shared_snaps, snap)) { dsl_dataset_t *ds = snap->ds; /* * Property callbacks are registered to a particular * dsl_dir. Since ours is changing, evict the objset * so that they will be unregistered from the old dsl_dir. */ if (ds->ds_objset) { dmu_objset_evict(ds->ds_objset); ds->ds_objset = NULL; } /* move snap name entry */ VERIFY0(dsl_dataset_get_snapname(ds)); VERIFY0(dsl_dataset_snap_remove(origin_head, ds->ds_snapname, tx, B_TRUE)); VERIFY0(zap_add(dp->dp_meta_objset, dsl_dataset_phys(hds)->ds_snapnames_zapobj, ds->ds_snapname, 8, 1, &ds->ds_object, tx)); dsl_fs_ss_count_adjust(hds->ds_dir, 1, DD_FIELD_SNAPSHOT_COUNT, tx); /* change containing dsl_dir */ dmu_buf_will_dirty(ds->ds_dbuf, tx); ASSERT3U(dsl_dataset_phys(ds)->ds_dir_obj, ==, odd->dd_object); dsl_dataset_phys(ds)->ds_dir_obj = dd->dd_object; ASSERT3P(ds->ds_dir, ==, odd); dsl_dir_rele(ds->ds_dir, ds); VERIFY0(dsl_dir_hold_obj(dp, dd->dd_object, NULL, ds, &ds->ds_dir)); /* move any clone references */ if (dsl_dataset_phys(ds)->ds_next_clones_obj && spa_version(dp->dp_spa) >= SPA_VERSION_DIR_CLONES) { zap_cursor_t zc; zap_attribute_t za; for (zap_cursor_init(&zc, dp->dp_meta_objset, dsl_dataset_phys(ds)->ds_next_clones_obj); zap_cursor_retrieve(&zc, &za) == 0; zap_cursor_advance(&zc)) { dsl_dataset_t *cnds; uint64_t o; if (za.za_first_integer == oldnext_obj) { /* * We've already moved the * origin's reference. */ continue; } VERIFY0(dsl_dataset_hold_obj(dp, za.za_first_integer, FTAG, &cnds)); o = dsl_dir_phys(cnds->ds_dir)-> dd_head_dataset_obj; VERIFY0(zap_remove_int(dp->dp_meta_objset, dsl_dir_phys(odd)->dd_clones, o, tx)); VERIFY0(zap_add_int(dp->dp_meta_objset, dsl_dir_phys(dd)->dd_clones, o, tx)); dsl_dataset_rele(cnds, FTAG); } zap_cursor_fini(&zc); } ASSERT(!dsl_prop_hascb(ds)); } /* * Change space accounting. * Note, pa->*usedsnap and dd_used_breakdown[SNAP] will either * both be valid, or both be 0 (resulting in delta == 0). This * is true for each of {clone,origin} independently. */ delta = ddpa->cloneusedsnap - dsl_dir_phys(dd)->dd_used_breakdown[DD_USED_SNAP]; ASSERT3S(delta, >=, 0); ASSERT3U(ddpa->used, >=, delta); dsl_dir_diduse_space(dd, DD_USED_SNAP, delta, 0, 0, tx); dsl_dir_diduse_space(dd, DD_USED_HEAD, ddpa->used - delta, ddpa->comp, ddpa->uncomp, tx); delta = ddpa->originusedsnap - dsl_dir_phys(odd)->dd_used_breakdown[DD_USED_SNAP]; ASSERT3S(delta, <=, 0); ASSERT3U(ddpa->used, >=, -delta); dsl_dir_diduse_space(odd, DD_USED_SNAP, delta, 0, 0, tx); dsl_dir_diduse_space(odd, DD_USED_HEAD, -ddpa->used - delta, -ddpa->comp, -ddpa->uncomp, tx); dsl_dataset_phys(origin_ds)->ds_unique_bytes = ddpa->unique; /* * Since livelists are specific to a clone's origin txg, they * are no longer accurate. Destroy the livelist from the clone being * promoted. If the origin dataset is a clone, destroy its livelist * as well. */ dsl_dir_remove_livelist(dd, tx, B_TRUE); dsl_dir_remove_livelist(odd, tx, B_TRUE); /* log history record */ spa_history_log_internal_ds(hds, "promote", tx, " "); dsl_dir_rele(odd, FTAG); /* * Transfer common error blocks from old head to new head, before * calling promote_rele() on ddpa since we need to dereference * origin_head and hds. */ if (spa_feature_is_enabled(dp->dp_spa, SPA_FEATURE_HEAD_ERRLOG)) { uint64_t old_head = origin_head->ds_object; uint64_t new_head = hds->ds_object; spa_swap_errlog(dp->dp_spa, new_head, old_head, tx); } promote_rele(ddpa, FTAG); } /* * Make a list of dsl_dataset_t's for the snapshots between first_obj * (exclusive) and last_obj (inclusive). The list will be in reverse * order (last_obj will be the list_head()). If first_obj == 0, do all * snapshots back to this dataset's origin. */ static int snaplist_make(dsl_pool_t *dp, uint64_t first_obj, uint64_t last_obj, list_t *l, const void *tag) { uint64_t obj = last_obj; list_create(l, sizeof (struct promotenode), offsetof(struct promotenode, link)); while (obj != first_obj) { dsl_dataset_t *ds; struct promotenode *snap; int err; err = dsl_dataset_hold_obj(dp, obj, tag, &ds); ASSERT(err != ENOENT); if (err != 0) return (err); if (first_obj == 0) first_obj = dsl_dir_phys(ds->ds_dir)->dd_origin_obj; snap = kmem_alloc(sizeof (*snap), KM_SLEEP); snap->ds = ds; list_insert_tail(l, snap); obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; } return (0); } static int snaplist_space(list_t *l, uint64_t mintxg, uint64_t *spacep) { struct promotenode *snap; *spacep = 0; for (snap = list_head(l); snap; snap = list_next(l, snap)) { uint64_t used, comp, uncomp; dsl_deadlist_space_range(&snap->ds->ds_deadlist, mintxg, UINT64_MAX, &used, &comp, &uncomp); *spacep += used; } return (0); } static void snaplist_destroy(list_t *l, const void *tag) { struct promotenode *snap; if (l == NULL || !list_link_active(&l->list_head)) return; while ((snap = list_remove_tail(l)) != NULL) { dsl_dataset_rele(snap->ds, tag); kmem_free(snap, sizeof (*snap)); } list_destroy(l); } static int promote_hold(dsl_dataset_promote_arg_t *ddpa, dsl_pool_t *dp, const void *tag) { int error; dsl_dir_t *dd; struct promotenode *snap; error = dsl_dataset_hold(dp, ddpa->ddpa_clonename, tag, &ddpa->ddpa_clone); if (error != 0) return (error); dd = ddpa->ddpa_clone->ds_dir; if (ddpa->ddpa_clone->ds_is_snapshot || !dsl_dir_is_clone(dd)) { dsl_dataset_rele(ddpa->ddpa_clone, tag); return (SET_ERROR(EINVAL)); } error = snaplist_make(dp, 0, dsl_dir_phys(dd)->dd_origin_obj, &ddpa->shared_snaps, tag); if (error != 0) goto out; error = snaplist_make(dp, 0, ddpa->ddpa_clone->ds_object, &ddpa->clone_snaps, tag); if (error != 0) goto out; snap = list_head(&ddpa->shared_snaps); ASSERT3U(snap->ds->ds_object, ==, dsl_dir_phys(dd)->dd_origin_obj); error = snaplist_make(dp, dsl_dir_phys(dd)->dd_origin_obj, dsl_dir_phys(snap->ds->ds_dir)->dd_head_dataset_obj, &ddpa->origin_snaps, tag); if (error != 0) goto out; if (dsl_dir_phys(snap->ds->ds_dir)->dd_origin_obj != 0) { error = dsl_dataset_hold_obj(dp, dsl_dir_phys(snap->ds->ds_dir)->dd_origin_obj, tag, &ddpa->origin_origin); if (error != 0) goto out; } out: if (error != 0) promote_rele(ddpa, tag); return (error); } static void promote_rele(dsl_dataset_promote_arg_t *ddpa, const void *tag) { snaplist_destroy(&ddpa->shared_snaps, tag); snaplist_destroy(&ddpa->clone_snaps, tag); snaplist_destroy(&ddpa->origin_snaps, tag); if (ddpa->origin_origin != NULL) dsl_dataset_rele(ddpa->origin_origin, tag); dsl_dataset_rele(ddpa->ddpa_clone, tag); } /* * Promote a clone. * * If it fails due to a conflicting snapshot name, "conflsnap" will be filled * in with the name. (It must be at least ZFS_MAX_DATASET_NAME_LEN bytes long.) */ int dsl_dataset_promote(const char *name, char *conflsnap) { dsl_dataset_promote_arg_t ddpa = { 0 }; uint64_t numsnaps; int error; nvpair_t *snap_pair; objset_t *os; /* * We will modify space proportional to the number of * snapshots. Compute numsnaps. */ error = dmu_objset_hold(name, FTAG, &os); if (error != 0) return (error); error = zap_count(dmu_objset_pool(os)->dp_meta_objset, dsl_dataset_phys(dmu_objset_ds(os))->ds_snapnames_zapobj, &numsnaps); dmu_objset_rele(os, FTAG); if (error != 0) return (error); ddpa.ddpa_clonename = name; ddpa.err_ds = fnvlist_alloc(); ddpa.cr = CRED(); ddpa.proc = curproc; error = dsl_sync_task(name, dsl_dataset_promote_check, dsl_dataset_promote_sync, &ddpa, 2 + numsnaps, ZFS_SPACE_CHECK_RESERVED); /* * Return the first conflicting snapshot found. */ snap_pair = nvlist_next_nvpair(ddpa.err_ds, NULL); if (snap_pair != NULL && conflsnap != NULL) (void) strlcpy(conflsnap, nvpair_name(snap_pair), ZFS_MAX_DATASET_NAME_LEN); fnvlist_free(ddpa.err_ds); return (error); } int dsl_dataset_clone_swap_check_impl(dsl_dataset_t *clone, dsl_dataset_t *origin_head, boolean_t force, void *owner, dmu_tx_t *tx) { /* * "slack" factor for received datasets with refquota set on them. * See the bottom of this function for details on its use. */ uint64_t refquota_slack = (uint64_t)DMU_MAX_ACCESS * spa_asize_inflation; int64_t unused_refres_delta; /* they should both be heads */ if (clone->ds_is_snapshot || origin_head->ds_is_snapshot) return (SET_ERROR(EINVAL)); /* if we are not forcing, the branch point should be just before them */ if (!force && clone->ds_prev != origin_head->ds_prev) return (SET_ERROR(EINVAL)); /* clone should be the clone (unless they are unrelated) */ if (clone->ds_prev != NULL && clone->ds_prev != clone->ds_dir->dd_pool->dp_origin_snap && origin_head->ds_dir != clone->ds_prev->ds_dir) return (SET_ERROR(EINVAL)); /* the clone should be a child of the origin */ if (clone->ds_dir->dd_parent != origin_head->ds_dir) return (SET_ERROR(EINVAL)); /* origin_head shouldn't be modified unless 'force' */ if (!force && dsl_dataset_modified_since_snap(origin_head, origin_head->ds_prev)) return (SET_ERROR(ETXTBSY)); /* origin_head should have no long holds (e.g. is not mounted) */ if (dsl_dataset_handoff_check(origin_head, owner, tx)) return (SET_ERROR(EBUSY)); /* check amount of any unconsumed refreservation */ unused_refres_delta = (int64_t)MIN(origin_head->ds_reserved, dsl_dataset_phys(origin_head)->ds_unique_bytes) - (int64_t)MIN(origin_head->ds_reserved, dsl_dataset_phys(clone)->ds_unique_bytes); if (unused_refres_delta > 0 && unused_refres_delta > dsl_dir_space_available(origin_head->ds_dir, NULL, 0, TRUE)) return (SET_ERROR(ENOSPC)); /* * The clone can't be too much over the head's refquota. * * To ensure that the entire refquota can be used, we allow one * transaction to exceed the refquota. Therefore, this check * needs to also allow for the space referenced to be more than the * refquota. The maximum amount of space that one transaction can use * on disk is DMU_MAX_ACCESS * spa_asize_inflation. Allowing this * overage ensures that we are able to receive a filesystem that * exceeds the refquota on the source system. * * So that overage is the refquota_slack we use below. */ if (origin_head->ds_quota != 0 && dsl_dataset_phys(clone)->ds_referenced_bytes > origin_head->ds_quota + refquota_slack) return (SET_ERROR(EDQUOT)); return (0); } static void dsl_dataset_swap_remap_deadlists(dsl_dataset_t *clone, dsl_dataset_t *origin, dmu_tx_t *tx) { uint64_t clone_remap_dl_obj, origin_remap_dl_obj; dsl_pool_t *dp = dmu_tx_pool(tx); ASSERT(dsl_pool_sync_context(dp)); clone_remap_dl_obj = dsl_dataset_get_remap_deadlist_object(clone); origin_remap_dl_obj = dsl_dataset_get_remap_deadlist_object(origin); if (clone_remap_dl_obj != 0) { dsl_deadlist_close(&clone->ds_remap_deadlist); dsl_dataset_unset_remap_deadlist_object(clone, tx); } if (origin_remap_dl_obj != 0) { dsl_deadlist_close(&origin->ds_remap_deadlist); dsl_dataset_unset_remap_deadlist_object(origin, tx); } if (clone_remap_dl_obj != 0) { dsl_dataset_set_remap_deadlist_object(origin, clone_remap_dl_obj, tx); dsl_deadlist_open(&origin->ds_remap_deadlist, dp->dp_meta_objset, clone_remap_dl_obj); } if (origin_remap_dl_obj != 0) { dsl_dataset_set_remap_deadlist_object(clone, origin_remap_dl_obj, tx); dsl_deadlist_open(&clone->ds_remap_deadlist, dp->dp_meta_objset, origin_remap_dl_obj); } } void dsl_dataset_clone_swap_sync_impl(dsl_dataset_t *clone, dsl_dataset_t *origin_head, dmu_tx_t *tx) { dsl_pool_t *dp = dmu_tx_pool(tx); int64_t unused_refres_delta; ASSERT(clone->ds_reserved == 0); /* * NOTE: On DEBUG kernels there could be a race between this and * the check function if spa_asize_inflation is adjusted... */ ASSERT(origin_head->ds_quota == 0 || dsl_dataset_phys(clone)->ds_unique_bytes <= origin_head->ds_quota + DMU_MAX_ACCESS * spa_asize_inflation); ASSERT3P(clone->ds_prev, ==, origin_head->ds_prev); dsl_dir_cancel_waiters(origin_head->ds_dir); /* * Swap per-dataset feature flags. */ for (spa_feature_t f = 0; f < SPA_FEATURES; f++) { if (!(spa_feature_table[f].fi_flags & ZFEATURE_FLAG_PER_DATASET)) { ASSERT(!dsl_dataset_feature_is_active(clone, f)); ASSERT(!dsl_dataset_feature_is_active(origin_head, f)); continue; } boolean_t clone_inuse = dsl_dataset_feature_is_active(clone, f); void *clone_feature = clone->ds_feature[f]; boolean_t origin_head_inuse = dsl_dataset_feature_is_active(origin_head, f); void *origin_head_feature = origin_head->ds_feature[f]; if (clone_inuse) dsl_dataset_deactivate_feature_impl(clone, f, tx); if (origin_head_inuse) dsl_dataset_deactivate_feature_impl(origin_head, f, tx); if (clone_inuse) { dsl_dataset_activate_feature(origin_head->ds_object, f, clone_feature, tx); origin_head->ds_feature[f] = clone_feature; } if (origin_head_inuse) { dsl_dataset_activate_feature(clone->ds_object, f, origin_head_feature, tx); clone->ds_feature[f] = origin_head_feature; } } dmu_buf_will_dirty(clone->ds_dbuf, tx); dmu_buf_will_dirty(origin_head->ds_dbuf, tx); if (clone->ds_objset != NULL) { dmu_objset_evict(clone->ds_objset); clone->ds_objset = NULL; } if (origin_head->ds_objset != NULL) { dmu_objset_evict(origin_head->ds_objset); origin_head->ds_objset = NULL; } unused_refres_delta = (int64_t)MIN(origin_head->ds_reserved, dsl_dataset_phys(origin_head)->ds_unique_bytes) - (int64_t)MIN(origin_head->ds_reserved, dsl_dataset_phys(clone)->ds_unique_bytes); /* * Reset origin's unique bytes. */ { dsl_dataset_t *origin = clone->ds_prev; uint64_t comp, uncomp; dmu_buf_will_dirty(origin->ds_dbuf, tx); dsl_deadlist_space_range(&clone->ds_deadlist, dsl_dataset_phys(origin)->ds_prev_snap_txg, UINT64_MAX, &dsl_dataset_phys(origin)->ds_unique_bytes, &comp, &uncomp); } /* swap blkptrs */ { rrw_enter(&clone->ds_bp_rwlock, RW_WRITER, FTAG); rrw_enter(&origin_head->ds_bp_rwlock, RW_WRITER, FTAG); blkptr_t tmp; tmp = dsl_dataset_phys(origin_head)->ds_bp; dsl_dataset_phys(origin_head)->ds_bp = dsl_dataset_phys(clone)->ds_bp; dsl_dataset_phys(clone)->ds_bp = tmp; rrw_exit(&origin_head->ds_bp_rwlock, FTAG); rrw_exit(&clone->ds_bp_rwlock, FTAG); } /* set dd_*_bytes */ { int64_t dused, dcomp, duncomp; uint64_t cdl_used, cdl_comp, cdl_uncomp; uint64_t odl_used, odl_comp, odl_uncomp; ASSERT3U(dsl_dir_phys(clone->ds_dir)-> dd_used_breakdown[DD_USED_SNAP], ==, 0); dsl_deadlist_space(&clone->ds_deadlist, &cdl_used, &cdl_comp, &cdl_uncomp); dsl_deadlist_space(&origin_head->ds_deadlist, &odl_used, &odl_comp, &odl_uncomp); dused = dsl_dataset_phys(clone)->ds_referenced_bytes + cdl_used - (dsl_dataset_phys(origin_head)->ds_referenced_bytes + odl_used); dcomp = dsl_dataset_phys(clone)->ds_compressed_bytes + cdl_comp - (dsl_dataset_phys(origin_head)->ds_compressed_bytes + odl_comp); duncomp = dsl_dataset_phys(clone)->ds_uncompressed_bytes + cdl_uncomp - (dsl_dataset_phys(origin_head)->ds_uncompressed_bytes + odl_uncomp); dsl_dir_diduse_space(origin_head->ds_dir, DD_USED_HEAD, dused, dcomp, duncomp, tx); dsl_dir_diduse_space(clone->ds_dir, DD_USED_HEAD, -dused, -dcomp, -duncomp, tx); /* * The difference in the space used by snapshots is the * difference in snapshot space due to the head's * deadlist (since that's the only thing that's * changing that affects the snapused). */ dsl_deadlist_space_range(&clone->ds_deadlist, origin_head->ds_dir->dd_origin_txg, UINT64_MAX, &cdl_used, &cdl_comp, &cdl_uncomp); dsl_deadlist_space_range(&origin_head->ds_deadlist, origin_head->ds_dir->dd_origin_txg, UINT64_MAX, &odl_used, &odl_comp, &odl_uncomp); dsl_dir_transfer_space(origin_head->ds_dir, cdl_used - odl_used, DD_USED_HEAD, DD_USED_SNAP, tx); } /* swap ds_*_bytes */ SWITCH64(dsl_dataset_phys(origin_head)->ds_referenced_bytes, dsl_dataset_phys(clone)->ds_referenced_bytes); SWITCH64(dsl_dataset_phys(origin_head)->ds_compressed_bytes, dsl_dataset_phys(clone)->ds_compressed_bytes); SWITCH64(dsl_dataset_phys(origin_head)->ds_uncompressed_bytes, dsl_dataset_phys(clone)->ds_uncompressed_bytes); SWITCH64(dsl_dataset_phys(origin_head)->ds_unique_bytes, dsl_dataset_phys(clone)->ds_unique_bytes); /* apply any parent delta for change in unconsumed refreservation */ dsl_dir_diduse_space(origin_head->ds_dir, DD_USED_REFRSRV, unused_refres_delta, 0, 0, tx); /* * Swap deadlists. */ dsl_deadlist_close(&clone->ds_deadlist); dsl_deadlist_close(&origin_head->ds_deadlist); SWITCH64(dsl_dataset_phys(origin_head)->ds_deadlist_obj, dsl_dataset_phys(clone)->ds_deadlist_obj); dsl_deadlist_open(&clone->ds_deadlist, dp->dp_meta_objset, dsl_dataset_phys(clone)->ds_deadlist_obj); dsl_deadlist_open(&origin_head->ds_deadlist, dp->dp_meta_objset, dsl_dataset_phys(origin_head)->ds_deadlist_obj); dsl_dataset_swap_remap_deadlists(clone, origin_head, tx); /* * If there is a bookmark at the origin, its "next dataset" is * changing, so we need to reset its FBN. */ dsl_bookmark_next_changed(origin_head, origin_head->ds_prev, tx); dsl_scan_ds_clone_swapped(origin_head, clone, tx); /* * Destroy any livelists associated with the clone or the origin, * since after the swap the corresponding livelists are no longer * valid. */ dsl_dir_remove_livelist(clone->ds_dir, tx, B_TRUE); dsl_dir_remove_livelist(origin_head->ds_dir, tx, B_TRUE); spa_history_log_internal_ds(clone, "clone swap", tx, "parent=%s", origin_head->ds_dir->dd_myname); } /* * Given a pool name and a dataset object number in that pool, * return the name of that dataset. */ int dsl_dsobj_to_dsname(char *pname, uint64_t obj, char *buf) { dsl_pool_t *dp; dsl_dataset_t *ds; int error; error = dsl_pool_hold(pname, FTAG, &dp); if (error != 0) return (error); error = dsl_dataset_hold_obj(dp, obj, FTAG, &ds); if (error == 0) { dsl_dataset_name(ds, buf); dsl_dataset_rele(ds, FTAG); } dsl_pool_rele(dp, FTAG); return (error); } int dsl_dataset_check_quota(dsl_dataset_t *ds, boolean_t check_quota, uint64_t asize, uint64_t inflight, uint64_t *used, uint64_t *ref_rsrv) { int error = 0; ASSERT3S(asize, >, 0); /* * *ref_rsrv is the portion of asize that will come from any * unconsumed refreservation space. */ *ref_rsrv = 0; mutex_enter(&ds->ds_lock); /* * Make a space adjustment for reserved bytes. */ if (ds->ds_reserved > dsl_dataset_phys(ds)->ds_unique_bytes) { ASSERT3U(*used, >=, ds->ds_reserved - dsl_dataset_phys(ds)->ds_unique_bytes); *used -= (ds->ds_reserved - dsl_dataset_phys(ds)->ds_unique_bytes); *ref_rsrv = asize - MIN(asize, parent_delta(ds, asize + inflight)); } if (!check_quota || ds->ds_quota == 0) { mutex_exit(&ds->ds_lock); return (0); } /* * If they are requesting more space, and our current estimate * is over quota, they get to try again unless the actual * on-disk is over quota and there are no pending changes (which * may free up space for us). */ if (dsl_dataset_phys(ds)->ds_referenced_bytes + inflight >= ds->ds_quota) { if (inflight > 0 || dsl_dataset_phys(ds)->ds_referenced_bytes < ds->ds_quota) error = SET_ERROR(ERESTART); else error = SET_ERROR(EDQUOT); } mutex_exit(&ds->ds_lock); return (error); } typedef struct dsl_dataset_set_qr_arg { const char *ddsqra_name; zprop_source_t ddsqra_source; uint64_t ddsqra_value; } dsl_dataset_set_qr_arg_t; static int dsl_dataset_set_refquota_check(void *arg, dmu_tx_t *tx) { dsl_dataset_set_qr_arg_t *ddsqra = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *ds; int error; uint64_t newval; if (spa_version(dp->dp_spa) < SPA_VERSION_REFQUOTA) return (SET_ERROR(ENOTSUP)); error = dsl_dataset_hold(dp, ddsqra->ddsqra_name, FTAG, &ds); if (error != 0) return (error); if (ds->ds_is_snapshot) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(EINVAL)); } error = dsl_prop_predict(ds->ds_dir, zfs_prop_to_name(ZFS_PROP_REFQUOTA), ddsqra->ddsqra_source, ddsqra->ddsqra_value, &newval); if (error != 0) { dsl_dataset_rele(ds, FTAG); return (error); } if (newval == 0) { dsl_dataset_rele(ds, FTAG); return (0); } if (newval < dsl_dataset_phys(ds)->ds_referenced_bytes || newval < ds->ds_reserved) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(ENOSPC)); } dsl_dataset_rele(ds, FTAG); return (0); } static void dsl_dataset_set_refquota_sync(void *arg, dmu_tx_t *tx) { dsl_dataset_set_qr_arg_t *ddsqra = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *ds = NULL; uint64_t newval; VERIFY0(dsl_dataset_hold(dp, ddsqra->ddsqra_name, FTAG, &ds)); dsl_prop_set_sync_impl(ds, zfs_prop_to_name(ZFS_PROP_REFQUOTA), ddsqra->ddsqra_source, sizeof (ddsqra->ddsqra_value), 1, &ddsqra->ddsqra_value, tx); VERIFY0(dsl_prop_get_int_ds(ds, zfs_prop_to_name(ZFS_PROP_REFQUOTA), &newval)); if (ds->ds_quota != newval) { dmu_buf_will_dirty(ds->ds_dbuf, tx); ds->ds_quota = newval; } dsl_dataset_rele(ds, FTAG); } int dsl_dataset_set_refquota(const char *dsname, zprop_source_t source, uint64_t refquota) { dsl_dataset_set_qr_arg_t ddsqra; ddsqra.ddsqra_name = dsname; ddsqra.ddsqra_source = source; ddsqra.ddsqra_value = refquota; return (dsl_sync_task(dsname, dsl_dataset_set_refquota_check, dsl_dataset_set_refquota_sync, &ddsqra, 0, ZFS_SPACE_CHECK_EXTRA_RESERVED)); } static int dsl_dataset_set_refreservation_check(void *arg, dmu_tx_t *tx) { dsl_dataset_set_qr_arg_t *ddsqra = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *ds; int error; uint64_t newval, unique; if (spa_version(dp->dp_spa) < SPA_VERSION_REFRESERVATION) return (SET_ERROR(ENOTSUP)); error = dsl_dataset_hold(dp, ddsqra->ddsqra_name, FTAG, &ds); if (error != 0) return (error); if (ds->ds_is_snapshot) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(EINVAL)); } error = dsl_prop_predict(ds->ds_dir, zfs_prop_to_name(ZFS_PROP_REFRESERVATION), ddsqra->ddsqra_source, ddsqra->ddsqra_value, &newval); if (error != 0) { dsl_dataset_rele(ds, FTAG); return (error); } /* * If we are doing the preliminary check in open context, the * space estimates may be inaccurate. */ if (!dmu_tx_is_syncing(tx)) { dsl_dataset_rele(ds, FTAG); return (0); } mutex_enter(&ds->ds_lock); if (!DS_UNIQUE_IS_ACCURATE(ds)) dsl_dataset_recalc_head_uniq(ds); unique = dsl_dataset_phys(ds)->ds_unique_bytes; mutex_exit(&ds->ds_lock); if (MAX(unique, newval) > MAX(unique, ds->ds_reserved)) { uint64_t delta = MAX(unique, newval) - MAX(unique, ds->ds_reserved); if (delta > dsl_dir_space_available(ds->ds_dir, NULL, 0, B_TRUE) || (ds->ds_quota > 0 && newval > ds->ds_quota)) { dsl_dataset_rele(ds, FTAG); return (SET_ERROR(ENOSPC)); } } dsl_dataset_rele(ds, FTAG); return (0); } void dsl_dataset_set_refreservation_sync_impl(dsl_dataset_t *ds, zprop_source_t source, uint64_t value, dmu_tx_t *tx) { uint64_t newval; uint64_t unique; int64_t delta; dsl_prop_set_sync_impl(ds, zfs_prop_to_name(ZFS_PROP_REFRESERVATION), source, sizeof (value), 1, &value, tx); VERIFY0(dsl_prop_get_int_ds(ds, zfs_prop_to_name(ZFS_PROP_REFRESERVATION), &newval)); dmu_buf_will_dirty(ds->ds_dbuf, tx); mutex_enter(&ds->ds_dir->dd_lock); mutex_enter(&ds->ds_lock); ASSERT(DS_UNIQUE_IS_ACCURATE(ds)); unique = dsl_dataset_phys(ds)->ds_unique_bytes; delta = MAX(0, (int64_t)(newval - unique)) - MAX(0, (int64_t)(ds->ds_reserved - unique)); ds->ds_reserved = newval; mutex_exit(&ds->ds_lock); dsl_dir_diduse_space(ds->ds_dir, DD_USED_REFRSRV, delta, 0, 0, tx); mutex_exit(&ds->ds_dir->dd_lock); } static void dsl_dataset_set_refreservation_sync(void *arg, dmu_tx_t *tx) { dsl_dataset_set_qr_arg_t *ddsqra = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *ds = NULL; VERIFY0(dsl_dataset_hold(dp, ddsqra->ddsqra_name, FTAG, &ds)); dsl_dataset_set_refreservation_sync_impl(ds, ddsqra->ddsqra_source, ddsqra->ddsqra_value, tx); dsl_dataset_rele(ds, FTAG); } int dsl_dataset_set_refreservation(const char *dsname, zprop_source_t source, uint64_t refreservation) { dsl_dataset_set_qr_arg_t ddsqra; ddsqra.ddsqra_name = dsname; ddsqra.ddsqra_source = source; ddsqra.ddsqra_value = refreservation; return (dsl_sync_task(dsname, dsl_dataset_set_refreservation_check, dsl_dataset_set_refreservation_sync, &ddsqra, 0, ZFS_SPACE_CHECK_EXTRA_RESERVED)); } typedef struct dsl_dataset_set_compression_arg { const char *ddsca_name; zprop_source_t ddsca_source; uint64_t ddsca_value; } dsl_dataset_set_compression_arg_t; static int dsl_dataset_set_compression_check(void *arg, dmu_tx_t *tx) { dsl_dataset_set_compression_arg_t *ddsca = arg; dsl_pool_t *dp = dmu_tx_pool(tx); uint64_t compval = ZIO_COMPRESS_ALGO(ddsca->ddsca_value); spa_feature_t f = zio_compress_to_feature(compval); if (f == SPA_FEATURE_NONE) return (SET_ERROR(EINVAL)); if (!spa_feature_is_enabled(dp->dp_spa, f)) return (SET_ERROR(ENOTSUP)); return (0); } static void dsl_dataset_set_compression_sync(void *arg, dmu_tx_t *tx) { dsl_dataset_set_compression_arg_t *ddsca = arg; dsl_pool_t *dp = dmu_tx_pool(tx); dsl_dataset_t *ds = NULL; uint64_t compval = ZIO_COMPRESS_ALGO(ddsca->ddsca_value); spa_feature_t f = zio_compress_to_feature(compval); ASSERT3S(f, !=, SPA_FEATURE_NONE); ASSERT3S(spa_feature_table[f].fi_type, ==, ZFEATURE_TYPE_BOOLEAN); VERIFY0(dsl_dataset_hold(dp, ddsca->ddsca_name, FTAG, &ds)); if (zfeature_active(f, ds->ds_feature[f]) != B_TRUE) { ds->ds_feature_activation[f] = (void *)B_TRUE; dsl_dataset_activate_feature(ds->ds_object, f, ds->ds_feature_activation[f], tx); ds->ds_feature[f] = ds->ds_feature_activation[f]; } dsl_dataset_rele(ds, FTAG); } int dsl_dataset_set_compression(const char *dsname, zprop_source_t source, uint64_t compression) { dsl_dataset_set_compression_arg_t ddsca; /* * The sync task is only required for zstd in order to activate * the feature flag when the property is first set. */ if (ZIO_COMPRESS_ALGO(compression) != ZIO_COMPRESS_ZSTD) return (0); ddsca.ddsca_name = dsname; ddsca.ddsca_source = source; ddsca.ddsca_value = compression; return (dsl_sync_task(dsname, dsl_dataset_set_compression_check, dsl_dataset_set_compression_sync, &ddsca, 0, ZFS_SPACE_CHECK_EXTRA_RESERVED)); } /* * Return (in *usedp) the amount of space referenced by "new" that was not * referenced at the time the bookmark corresponds to. "New" may be a * snapshot or a head. The bookmark must be before new, in * new's filesystem (or its origin) -- caller verifies this. * * The written space is calculated by considering two components: First, we * ignore any freed space, and calculate the written as new's used space * minus old's used space. Next, we add in the amount of space that was freed * between the two time points, thus reducing new's used space relative to * old's. Specifically, this is the space that was born before * zbm_creation_txg, and freed before new (ie. on new's deadlist or a * previous deadlist). * * space freed [---------------------] * snapshots ---O-------O--------O-------O------ * bookmark new * * Note, the bookmark's zbm_*_bytes_refd must be valid, but if the HAS_FBN * flag is not set, we will calculate the freed_before_next based on the * next snapshot's deadlist, rather than using zbm_*_freed_before_next_snap. */ static int dsl_dataset_space_written_impl(zfs_bookmark_phys_t *bmp, dsl_dataset_t *new, uint64_t *usedp, uint64_t *compp, uint64_t *uncompp) { int err = 0; dsl_pool_t *dp = new->ds_dir->dd_pool; ASSERT(dsl_pool_config_held(dp)); if (dsl_dataset_is_snapshot(new)) { ASSERT3U(bmp->zbm_creation_txg, <, dsl_dataset_phys(new)->ds_creation_txg); } *usedp = 0; *usedp += dsl_dataset_phys(new)->ds_referenced_bytes; *usedp -= bmp->zbm_referenced_bytes_refd; *compp = 0; *compp += dsl_dataset_phys(new)->ds_compressed_bytes; *compp -= bmp->zbm_compressed_bytes_refd; *uncompp = 0; *uncompp += dsl_dataset_phys(new)->ds_uncompressed_bytes; *uncompp -= bmp->zbm_uncompressed_bytes_refd; dsl_dataset_t *snap = new; while (dsl_dataset_phys(snap)->ds_prev_snap_txg > bmp->zbm_creation_txg) { uint64_t used, comp, uncomp; dsl_deadlist_space_range(&snap->ds_deadlist, 0, bmp->zbm_creation_txg, &used, &comp, &uncomp); *usedp += used; *compp += comp; *uncompp += uncomp; uint64_t snapobj = dsl_dataset_phys(snap)->ds_prev_snap_obj; if (snap != new) dsl_dataset_rele(snap, FTAG); err = dsl_dataset_hold_obj(dp, snapobj, FTAG, &snap); if (err != 0) break; } /* * We might not have the FBN if we are calculating written from * a snapshot (because we didn't know the correct "next" snapshot * until now). */ if (bmp->zbm_flags & ZBM_FLAG_HAS_FBN) { *usedp += bmp->zbm_referenced_freed_before_next_snap; *compp += bmp->zbm_compressed_freed_before_next_snap; *uncompp += bmp->zbm_uncompressed_freed_before_next_snap; } else { ASSERT3U(dsl_dataset_phys(snap)->ds_prev_snap_txg, ==, bmp->zbm_creation_txg); uint64_t used, comp, uncomp; dsl_deadlist_space(&snap->ds_deadlist, &used, &comp, &uncomp); *usedp += used; *compp += comp; *uncompp += uncomp; } if (snap != new) dsl_dataset_rele(snap, FTAG); return (err); } /* * Return (in *usedp) the amount of space written in new that was not * present at the time the bookmark corresponds to. New may be a * snapshot or the head. Old must be a bookmark before new, in * new's filesystem (or its origin) -- caller verifies this. */ int dsl_dataset_space_written_bookmark(zfs_bookmark_phys_t *bmp, dsl_dataset_t *new, uint64_t *usedp, uint64_t *compp, uint64_t *uncompp) { if (!(bmp->zbm_flags & ZBM_FLAG_HAS_FBN)) return (SET_ERROR(ENOTSUP)); return (dsl_dataset_space_written_impl(bmp, new, usedp, compp, uncompp)); } /* * Return (in *usedp) the amount of space written in new that is not * present in oldsnap. New may be a snapshot or the head. Old must be * a snapshot before new, in new's filesystem (or its origin). If not then * fail and return EINVAL. */ int dsl_dataset_space_written(dsl_dataset_t *oldsnap, dsl_dataset_t *new, uint64_t *usedp, uint64_t *compp, uint64_t *uncompp) { if (!dsl_dataset_is_before(new, oldsnap, 0)) return (SET_ERROR(EINVAL)); zfs_bookmark_phys_t zbm = { 0 }; dsl_dataset_phys_t *dsp = dsl_dataset_phys(oldsnap); zbm.zbm_guid = dsp->ds_guid; zbm.zbm_creation_txg = dsp->ds_creation_txg; zbm.zbm_creation_time = dsp->ds_creation_time; zbm.zbm_referenced_bytes_refd = dsp->ds_referenced_bytes; zbm.zbm_compressed_bytes_refd = dsp->ds_compressed_bytes; zbm.zbm_uncompressed_bytes_refd = dsp->ds_uncompressed_bytes; /* * If oldsnap is the origin (or origin's origin, ...) of new, * we can't easily calculate the effective FBN. Therefore, * we do not set ZBM_FLAG_HAS_FBN, so that the _impl will calculate * it relative to the correct "next": the next snapshot towards "new", * rather than the next snapshot in oldsnap's dsl_dir. */ return (dsl_dataset_space_written_impl(&zbm, new, usedp, compp, uncompp)); } /* * Return (in *usedp) the amount of space that will be reclaimed if firstsnap, * lastsnap, and all snapshots in between are deleted. * * blocks that would be freed [---------------------------] * snapshots ---O-------O--------O-------O--------O * firstsnap lastsnap * * This is the set of blocks that were born after the snap before firstsnap, * (birth > firstsnap->prev_snap_txg) and died before the snap after the * last snap (ie, is on lastsnap->ds_next->ds_deadlist or an earlier deadlist). * We calculate this by iterating over the relevant deadlists (from the snap * after lastsnap, backward to the snap after firstsnap), summing up the * space on the deadlist that was born after the snap before firstsnap. */ int dsl_dataset_space_wouldfree(dsl_dataset_t *firstsnap, dsl_dataset_t *lastsnap, uint64_t *usedp, uint64_t *compp, uint64_t *uncompp) { int err = 0; uint64_t snapobj; dsl_pool_t *dp = firstsnap->ds_dir->dd_pool; ASSERT(firstsnap->ds_is_snapshot); ASSERT(lastsnap->ds_is_snapshot); /* * Check that the snapshots are in the same dsl_dir, and firstsnap * is before lastsnap. */ if (firstsnap->ds_dir != lastsnap->ds_dir || dsl_dataset_phys(firstsnap)->ds_creation_txg > dsl_dataset_phys(lastsnap)->ds_creation_txg) return (SET_ERROR(EINVAL)); *usedp = *compp = *uncompp = 0; snapobj = dsl_dataset_phys(lastsnap)->ds_next_snap_obj; while (snapobj != firstsnap->ds_object) { dsl_dataset_t *ds; uint64_t used, comp, uncomp; err = dsl_dataset_hold_obj(dp, snapobj, FTAG, &ds); if (err != 0) break; dsl_deadlist_space_range(&ds->ds_deadlist, dsl_dataset_phys(firstsnap)->ds_prev_snap_txg, UINT64_MAX, &used, &comp, &uncomp); *usedp += used; *compp += comp; *uncompp += uncomp; snapobj = dsl_dataset_phys(ds)->ds_prev_snap_obj; ASSERT3U(snapobj, !=, 0); dsl_dataset_rele(ds, FTAG); } return (err); } /* * Return TRUE if 'earlier' is an earlier snapshot in 'later's timeline. * For example, they could both be snapshots of the same filesystem, and * 'earlier' is before 'later'. Or 'earlier' could be the origin of * 'later's filesystem. Or 'earlier' could be an older snapshot in the origin's * filesystem. Or 'earlier' could be the origin's origin. * * If non-zero, earlier_txg is used instead of earlier's ds_creation_txg. */ boolean_t dsl_dataset_is_before(dsl_dataset_t *later, dsl_dataset_t *earlier, uint64_t earlier_txg) { dsl_pool_t *dp = later->ds_dir->dd_pool; int error; boolean_t ret; ASSERT(dsl_pool_config_held(dp)); ASSERT(earlier->ds_is_snapshot || earlier_txg != 0); if (earlier_txg == 0) earlier_txg = dsl_dataset_phys(earlier)->ds_creation_txg; if (later->ds_is_snapshot && earlier_txg >= dsl_dataset_phys(later)->ds_creation_txg) return (B_FALSE); if (later->ds_dir == earlier->ds_dir) return (B_TRUE); /* * We check dd_origin_obj explicitly here rather than using * dsl_dir_is_clone() so that we will return TRUE if "earlier" * is $ORIGIN@$ORIGIN. dsl_dataset_space_written() depends on * this behavior. */ if (dsl_dir_phys(later->ds_dir)->dd_origin_obj == 0) return (B_FALSE); dsl_dataset_t *origin; error = dsl_dataset_hold_obj(dp, dsl_dir_phys(later->ds_dir)->dd_origin_obj, FTAG, &origin); if (error != 0) return (B_FALSE); if (dsl_dataset_phys(origin)->ds_creation_txg == earlier_txg && origin->ds_dir == earlier->ds_dir) { dsl_dataset_rele(origin, FTAG); return (B_TRUE); } ret = dsl_dataset_is_before(origin, earlier, earlier_txg); dsl_dataset_rele(origin, FTAG); return (ret); } void dsl_dataset_zapify(dsl_dataset_t *ds, dmu_tx_t *tx) { objset_t *mos = ds->ds_dir->dd_pool->dp_meta_objset; dmu_object_zapify(mos, ds->ds_object, DMU_OT_DSL_DATASET, tx); } boolean_t dsl_dataset_is_zapified(dsl_dataset_t *ds) { dmu_object_info_t doi; dmu_object_info_from_db(ds->ds_dbuf, &doi); return (doi.doi_type == DMU_OTN_ZAP_METADATA); } boolean_t dsl_dataset_has_resume_receive_state(dsl_dataset_t *ds) { return (dsl_dataset_is_zapified(ds) && zap_contains(ds->ds_dir->dd_pool->dp_meta_objset, ds->ds_object, DS_FIELD_RESUME_TOGUID) == 0); } uint64_t dsl_dataset_get_remap_deadlist_object(dsl_dataset_t *ds) { uint64_t remap_deadlist_obj; int err; if (!dsl_dataset_is_zapified(ds)) return (0); err = zap_lookup(ds->ds_dir->dd_pool->dp_meta_objset, ds->ds_object, DS_FIELD_REMAP_DEADLIST, sizeof (remap_deadlist_obj), 1, &remap_deadlist_obj); if (err != 0) { VERIFY3S(err, ==, ENOENT); return (0); } ASSERT(remap_deadlist_obj != 0); return (remap_deadlist_obj); } boolean_t dsl_dataset_remap_deadlist_exists(dsl_dataset_t *ds) { EQUIV(dsl_deadlist_is_open(&ds->ds_remap_deadlist), dsl_dataset_get_remap_deadlist_object(ds) != 0); return (dsl_deadlist_is_open(&ds->ds_remap_deadlist)); } static void dsl_dataset_set_remap_deadlist_object(dsl_dataset_t *ds, uint64_t obj, dmu_tx_t *tx) { ASSERT(obj != 0); dsl_dataset_zapify(ds, tx); VERIFY0(zap_add(ds->ds_dir->dd_pool->dp_meta_objset, ds->ds_object, DS_FIELD_REMAP_DEADLIST, sizeof (obj), 1, &obj, tx)); } static void dsl_dataset_unset_remap_deadlist_object(dsl_dataset_t *ds, dmu_tx_t *tx) { VERIFY0(zap_remove(ds->ds_dir->dd_pool->dp_meta_objset, ds->ds_object, DS_FIELD_REMAP_DEADLIST, tx)); } void dsl_dataset_destroy_remap_deadlist(dsl_dataset_t *ds, dmu_tx_t *tx) { uint64_t remap_deadlist_object; spa_t *spa = ds->ds_dir->dd_pool->dp_spa; ASSERT(dmu_tx_is_syncing(tx)); ASSERT(dsl_dataset_remap_deadlist_exists(ds)); remap_deadlist_object = ds->ds_remap_deadlist.dl_object; dsl_deadlist_close(&ds->ds_remap_deadlist); dsl_deadlist_free(spa_meta_objset(spa), remap_deadlist_object, tx); dsl_dataset_unset_remap_deadlist_object(ds, tx); spa_feature_decr(spa, SPA_FEATURE_OBSOLETE_COUNTS, tx); } void dsl_dataset_create_remap_deadlist(dsl_dataset_t *ds, dmu_tx_t *tx) { uint64_t remap_deadlist_obj; spa_t *spa = ds->ds_dir->dd_pool->dp_spa; ASSERT(dmu_tx_is_syncing(tx)); ASSERT(MUTEX_HELD(&ds->ds_remap_deadlist_lock)); /* * Currently we only create remap deadlists when there are indirect * vdevs with referenced mappings. */ ASSERT(spa_feature_is_active(spa, SPA_FEATURE_DEVICE_REMOVAL)); remap_deadlist_obj = dsl_deadlist_clone( &ds->ds_deadlist, UINT64_MAX, dsl_dataset_phys(ds)->ds_prev_snap_obj, tx); dsl_dataset_set_remap_deadlist_object(ds, remap_deadlist_obj, tx); dsl_deadlist_open(&ds->ds_remap_deadlist, spa_meta_objset(spa), remap_deadlist_obj); spa_feature_incr(spa, SPA_FEATURE_OBSOLETE_COUNTS, tx); } void dsl_dataset_activate_redaction(dsl_dataset_t *ds, uint64_t *redact_snaps, uint64_t num_redact_snaps, dmu_tx_t *tx) { uint64_t dsobj = ds->ds_object; struct feature_type_uint64_array_arg *ftuaa = kmem_zalloc(sizeof (*ftuaa), KM_SLEEP); ftuaa->length = (int64_t)num_redact_snaps; if (num_redact_snaps > 0) { ftuaa->array = kmem_alloc(num_redact_snaps * sizeof (uint64_t), KM_SLEEP); memcpy(ftuaa->array, redact_snaps, num_redact_snaps * sizeof (uint64_t)); } dsl_dataset_activate_feature(dsobj, SPA_FEATURE_REDACTED_DATASETS, ftuaa, tx); ds->ds_feature[SPA_FEATURE_REDACTED_DATASETS] = ftuaa; } /* * Find and return (in *oldest_dsobj) the oldest snapshot of the dsobj * dataset whose birth time is >= min_txg. */ int dsl_dataset_oldest_snapshot(spa_t *spa, uint64_t head_ds, uint64_t min_txg, uint64_t *oldest_dsobj) { dsl_dataset_t *ds; dsl_pool_t *dp = spa->spa_dsl_pool; int error = dsl_dataset_hold_obj(dp, head_ds, 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 && min_txg < prev_obj_txg) { dsl_dataset_rele(ds, FTAG); if ((error = dsl_dataset_hold_obj(dp, prev_obj, FTAG, &ds)) != 0) return (error); prev_obj_txg = dsl_dataset_phys(ds)->ds_prev_snap_txg; prev_obj = dsl_dataset_phys(ds)->ds_prev_snap_obj; } *oldest_dsobj = ds->ds_object; dsl_dataset_rele(ds, FTAG); return (0); } ZFS_MODULE_PARAM(zfs, zfs_, max_recordsize, UINT, ZMOD_RW, "Max allowed record size"); ZFS_MODULE_PARAM(zfs, zfs_, allow_redacted_dataset_mount, INT, ZMOD_RW, "Allow mounting of redacted datasets"); ZFS_MODULE_PARAM(zfs, zfs_, snapshot_history_enabled, INT, ZMOD_RW, "Include snapshot events in pool history/events"); EXPORT_SYMBOL(dsl_dataset_hold); EXPORT_SYMBOL(dsl_dataset_hold_flags); EXPORT_SYMBOL(dsl_dataset_hold_obj); EXPORT_SYMBOL(dsl_dataset_hold_obj_flags); EXPORT_SYMBOL(dsl_dataset_own); EXPORT_SYMBOL(dsl_dataset_own_obj); EXPORT_SYMBOL(dsl_dataset_name); EXPORT_SYMBOL(dsl_dataset_rele); EXPORT_SYMBOL(dsl_dataset_rele_flags); EXPORT_SYMBOL(dsl_dataset_disown); EXPORT_SYMBOL(dsl_dataset_tryown); EXPORT_SYMBOL(dsl_dataset_create_sync); EXPORT_SYMBOL(dsl_dataset_create_sync_dd); EXPORT_SYMBOL(dsl_dataset_snapshot_check); EXPORT_SYMBOL(dsl_dataset_snapshot_sync); EXPORT_SYMBOL(dsl_dataset_promote); EXPORT_SYMBOL(dsl_dataset_user_hold); EXPORT_SYMBOL(dsl_dataset_user_release); EXPORT_SYMBOL(dsl_dataset_get_holds); EXPORT_SYMBOL(dsl_dataset_get_blkptr); EXPORT_SYMBOL(dsl_dataset_get_spa); EXPORT_SYMBOL(dsl_dataset_modified_since_snap); EXPORT_SYMBOL(dsl_dataset_space_written); EXPORT_SYMBOL(dsl_dataset_space_wouldfree); EXPORT_SYMBOL(dsl_dataset_sync); EXPORT_SYMBOL(dsl_dataset_block_born); EXPORT_SYMBOL(dsl_dataset_block_kill); EXPORT_SYMBOL(dsl_dataset_dirty); EXPORT_SYMBOL(dsl_dataset_stats); EXPORT_SYMBOL(dsl_dataset_fast_stat); EXPORT_SYMBOL(dsl_dataset_space); EXPORT_SYMBOL(dsl_dataset_fsid_guid); EXPORT_SYMBOL(dsl_dsobj_to_dsname); EXPORT_SYMBOL(dsl_dataset_check_quota); EXPORT_SYMBOL(dsl_dataset_clone_swap_check_impl); EXPORT_SYMBOL(dsl_dataset_clone_swap_sync_impl); diff --git a/module/zfs/gzip.c b/module/zfs/gzip.c index 0ca66c2bd657..e7fd6f63c4be 100644 --- a/module/zfs/gzip.c +++ b/module/zfs/gzip.c @@ -1,107 +1,110 @@ /* * 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 2007 Sun Microsystems, Inc. All rights reserved. * Use is subject to license terms. */ #include #include #include #include #ifdef _KERNEL #include typedef size_t zlen_t; #define compress_func z_compress_level #define uncompress_func z_uncompress #else /* _KERNEL */ #include typedef uLongf zlen_t; #define compress_func compress2 #define uncompress_func uncompress #endif -size_t -zfs_gzip_compress(void *s_start, void *d_start, size_t s_len, +static size_t +zfs_gzip_compress_buf(void *s_start, void *d_start, size_t s_len, size_t d_len, int n) { int ret; zlen_t dstlen = d_len; ASSERT(d_len <= s_len); /* check if hardware accelerator can be used */ if (qat_dc_use_accel(s_len)) { ret = qat_compress(QAT_COMPRESS, s_start, s_len, d_start, d_len, &dstlen); if (ret == CPA_STATUS_SUCCESS) { return ((size_t)dstlen); } else if (ret == CPA_STATUS_INCOMPRESSIBLE) { if (d_len != s_len) return (s_len); memcpy(d_start, s_start, s_len); return (s_len); } /* if hardware compression fails, do it again with software */ } if (compress_func(d_start, &dstlen, s_start, s_len, n) != Z_OK) { if (d_len != s_len) return (s_len); memcpy(d_start, s_start, s_len); return (s_len); } return ((size_t)dstlen); } -int -zfs_gzip_decompress(void *s_start, void *d_start, size_t s_len, +static int +zfs_gzip_decompress_buf(void *s_start, void *d_start, size_t s_len, size_t d_len, int n) { (void) n; zlen_t dstlen = d_len; ASSERT(d_len >= s_len); /* check if hardware accelerator can be used */ if (qat_dc_use_accel(d_len)) { if (qat_compress(QAT_DECOMPRESS, s_start, s_len, d_start, d_len, &dstlen) == CPA_STATUS_SUCCESS) return (0); /* if hardware de-compress fail, do it again with software */ } if (uncompress_func(d_start, &dstlen, s_start, s_len) != Z_OK) return (-1); return (0); } + +ZFS_COMPRESS_WRAP_DECL(zfs_gzip_compress) +ZFS_DECOMPRESS_WRAP_DECL(zfs_gzip_decompress) diff --git a/module/zfs/lz4_zfs.c b/module/zfs/lz4_zfs.c index 698ed69956e3..0033b5e50d1a 100644 --- a/module/zfs/lz4_zfs.c +++ b/module/zfs/lz4_zfs.c @@ -1,881 +1,884 @@ /* * LZ4 - Fast LZ compression algorithm * Header File * Copyright (C) 2011-2013, Yann Collet. * BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php) * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are * met: * * * Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * * Redistributions in binary form must reproduce the above * copyright notice, this list of conditions and the following disclaimer * in the documentation and/or other materials provided with the * distribution. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * * You can contact the author at : * - LZ4 homepage : http://fastcompression.blogspot.com/p/lz4.html * - LZ4 source repository : http://code.google.com/p/lz4/ */ /* * N.B. - This file seems to be based on LZ4 r85, dated Dec 10, 2012 */ #include #include static int real_LZ4_compress(const char *source, char *dest, int isize, int osize); static int LZ4_compressCtx(void *ctx, const char *source, char *dest, int isize, int osize); static int LZ4_compress64kCtx(void *ctx, const char *source, char *dest, int isize, int osize); /* See lz4.c */ int LZ4_uncompress_unknownOutputSize(const char *source, char *dest, int isize, int maxOutputSize); static kmem_cache_t *lz4_cache; -size_t -zfs_lz4_compress(void *s_start, void *d_start, size_t s_len, +static size_t +zfs_lz4_compress_buf(void *s_start, void *d_start, size_t s_len, size_t d_len, int n) { (void) n; uint32_t bufsiz; char *dest = d_start; ASSERT(d_len >= sizeof (bufsiz)); bufsiz = real_LZ4_compress(s_start, &dest[sizeof (bufsiz)], s_len, d_len - sizeof (bufsiz)); /* Signal an error if the compression routine returned zero. */ if (bufsiz == 0) return (s_len); /* * The exact compressed size is needed by the decompression routine, * so it is stored at the start of the buffer. Note that this may be * less than the compressed block size, which is rounded up to a * multiple of 1< s_len) return (1); /* * Returns 0 on success (decompression function returned non-negative) * and non-zero on failure (decompression function returned negative). */ return (LZ4_uncompress_unknownOutputSize(&src[sizeof (bufsiz)], d_start, bufsiz, d_len) < 0); } +ZFS_COMPRESS_WRAP_DECL(zfs_lz4_compress) +ZFS_DECOMPRESS_WRAP_DECL(zfs_lz4_decompress) + /* * LZ4 API Description: * * Simple Functions: * real_LZ4_compress() : * isize : is the input size. Max supported value is ~1.9GB * return : the number of bytes written in buffer dest * or 0 if the compression fails (if LZ4_COMPRESSMIN is set). * note : destination buffer must be already allocated. * destination buffer must be sized to handle worst cases * situations (input data not compressible) worst case size * evaluation is provided by function LZ4_compressBound(). * * real_LZ4_uncompress() : * osize : is the output size, therefore the original size * return : the number of bytes read in the source buffer. * If the source stream is malformed, the function will stop * decoding and return a negative result, indicating the byte * position of the faulty instruction. This function never * writes beyond dest + osize, and is therefore protected * against malicious data packets. * note : destination buffer must be already allocated * note : real_LZ4_uncompress() is not used in ZFS so its code * is not present here. * * Advanced Functions * * LZ4_compressBound() : * Provides the maximum size that LZ4 may output in a "worst case" * scenario (input data not compressible) primarily useful for memory * allocation of output buffer. * * isize : is the input size. Max supported value is ~1.9GB * return : maximum output size in a "worst case" scenario * note : this function is limited by "int" range (2^31-1) * * LZ4_uncompress_unknownOutputSize() : * isize : is the input size, therefore the compressed size * maxOutputSize : is the size of the destination buffer (which must be * already allocated) * return : the number of bytes decoded in the destination buffer * (necessarily <= maxOutputSize). If the source stream is * malformed, the function will stop decoding and return a * negative result, indicating the byte position of the faulty * instruction. This function never writes beyond dest + * maxOutputSize, and is therefore protected against malicious * data packets. * note : Destination buffer must be already allocated. * This version is slightly slower than real_LZ4_uncompress() * * LZ4_compressCtx() : * This function explicitly handles the CTX memory structure. * * ILLUMOS CHANGES: the CTX memory structure must be explicitly allocated * by the caller (either on the stack or using kmem_cache_alloc). Passing * NULL isn't valid. * * LZ4_compress64kCtx() : * Same as LZ4_compressCtx(), but specific to small inputs (<64KB). * isize *Must* be <64KB, otherwise the output will be corrupted. * * ILLUMOS CHANGES: the CTX memory structure must be explicitly allocated * by the caller (either on the stack or using kmem_cache_alloc). Passing * NULL isn't valid. */ /* * Tuning parameters */ /* * COMPRESSIONLEVEL: Increasing this value improves compression ratio * Lowering this value reduces memory usage. Reduced memory usage * typically improves speed, due to cache effect (ex: L1 32KB for Intel, * L1 64KB for AMD). Memory usage formula : N->2^(N+2) Bytes * (examples : 12 -> 16KB ; 17 -> 512KB) */ #define COMPRESSIONLEVEL 12 /* * NOTCOMPRESSIBLE_CONFIRMATION: Decreasing this value will make the * algorithm skip faster data segments considered "incompressible". * This may decrease compression ratio dramatically, but will be * faster on incompressible data. Increasing this value will make * the algorithm search more before declaring a segment "incompressible". * This could improve compression a bit, but will be slower on * incompressible data. The default value (6) is recommended. */ #define NOTCOMPRESSIBLE_CONFIRMATION 6 /* * BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE: This will provide a boost to * performance for big endian cpu, but the resulting compressed stream * will be incompatible with little-endian CPU. You can set this option * to 1 in situations where data will stay within closed environment. * This option is useless on Little_Endian CPU (such as x86). */ /* #define BIG_ENDIAN_NATIVE_BUT_INCOMPATIBLE 1 */ /* * CPU Feature Detection */ /* 32 or 64 bits ? */ #if defined(_LP64) #define LZ4_ARCH64 1 #else #define LZ4_ARCH64 0 #endif /* * Little Endian or Big Endian? * Note: overwrite the below #define if you know your architecture endianness. */ #if defined(_ZFS_BIG_ENDIAN) #define LZ4_BIG_ENDIAN 1 #else /* * Little Endian assumed. PDP Endian and other very rare endian format * are unsupported. */ #undef LZ4_BIG_ENDIAN #endif /* * Unaligned memory access is automatically enabled for "common" CPU, * such as x86. For others CPU, the compiler will be more cautious, and * insert extra code to ensure aligned access is respected. If you know * your target CPU supports unaligned memory access, you may want to * force this option manually to improve performance */ #if defined(__ARM_FEATURE_UNALIGNED) #define LZ4_FORCE_UNALIGNED_ACCESS 1 #endif /* * Illumos : we can't use GCC's __builtin_ctz family of builtins in the * kernel * Linux : we can use GCC's __builtin_ctz family of builtins in the * kernel */ #undef LZ4_FORCE_SW_BITCOUNT #if defined(__sparc) #define LZ4_FORCE_SW_BITCOUNT #endif /* * Compiler Options */ /* Disable restrict */ #define restrict /* * Linux : GCC_VERSION is defined as of 3.9-rc1, so undefine it. * torvalds/linux@3f3f8d2f48acfd8ed3b8e6b7377935da57b27b16 */ #ifdef GCC_VERSION #undef GCC_VERSION #endif #define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__) #if (GCC_VERSION >= 302) || (__INTEL_COMPILER >= 800) || defined(__clang__) #define expect(expr, value) (__builtin_expect((expr), (value))) #else #define expect(expr, value) (expr) #endif #ifndef likely #define likely(expr) expect((expr) != 0, 1) #endif #ifndef unlikely #define unlikely(expr) expect((expr) != 0, 0) #endif #define lz4_bswap16(x) ((unsigned short int) ((((x) >> 8) & 0xffu) | \ (((x) & 0xffu) << 8))) /* Basic types */ #define BYTE uint8_t #define U16 uint16_t #define U32 uint32_t #define S32 int32_t #define U64 uint64_t #ifndef LZ4_FORCE_UNALIGNED_ACCESS #pragma pack(1) #endif typedef struct _U16_S { U16 v; } U16_S; typedef struct _U32_S { U32 v; } U32_S; typedef struct _U64_S { U64 v; } U64_S; #ifndef LZ4_FORCE_UNALIGNED_ACCESS #pragma pack() #endif #define A64(x) (((U64_S *)(x))->v) #define A32(x) (((U32_S *)(x))->v) #define A16(x) (((U16_S *)(x))->v) /* * Constants */ #define MINMATCH 4 #define HASH_LOG COMPRESSIONLEVEL #define HASHTABLESIZE (1 << HASH_LOG) #define HASH_MASK (HASHTABLESIZE - 1) #define SKIPSTRENGTH (NOTCOMPRESSIBLE_CONFIRMATION > 2 ? \ NOTCOMPRESSIBLE_CONFIRMATION : 2) #define COPYLENGTH 8 #define LASTLITERALS 5 #define MFLIMIT (COPYLENGTH + MINMATCH) #define MINLENGTH (MFLIMIT + 1) #define MAXD_LOG 16 #define MAX_DISTANCE ((1 << MAXD_LOG) - 1) #define ML_BITS 4 #define ML_MASK ((1U<> ((MINMATCH * 8) - \ HASH_LOG)) #define LZ4_HASH_VALUE(p) LZ4_HASH_FUNCTION(A32(p)) #define LZ4_WILDCOPY(s, d, e) do { LZ4_COPYPACKET(s, d) } while (d < e); #define LZ4_BLINDCOPY(s, d, l) { BYTE* e = (d) + l; LZ4_WILDCOPY(s, d, e); \ d = e; } /* Private functions */ #if LZ4_ARCH64 static inline int LZ4_NbCommonBytes(register U64 val) { #if defined(LZ4_BIG_ENDIAN) #if ((defined(__GNUC__) && (GCC_VERSION >= 304)) || defined(__clang__)) && \ !defined(LZ4_FORCE_SW_BITCOUNT) return (__builtin_clzll(val) >> 3); #else int r; if (!(val >> 32)) { r = 4; } else { r = 0; val >>= 32; } if (!(val >> 16)) { r += 2; val >>= 8; } else { val >>= 24; } r += (!val); return (r); #endif #else #if ((defined(__GNUC__) && (GCC_VERSION >= 304)) || defined(__clang__)) && \ !defined(LZ4_FORCE_SW_BITCOUNT) return (__builtin_ctzll(val) >> 3); #else static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2, 0, 3, 1, 3, 1, 4, 2, 7, 0, 2, 3, 6, 1, 5, 3, 5, 1, 3, 4, 4, 2, 5, 6, 7, 7, 0, 1, 2, 3, 3, 4, 6, 2, 6, 5, 5, 3, 4, 5, 6, 7, 1, 2, 4, 6, 4, 4, 5, 7, 2, 6, 5, 7, 6, 7, 7 }; return DeBruijnBytePos[((U64) ((val & -val) * 0x0218A392CDABBD3F)) >> 58]; #endif #endif } #else static inline int LZ4_NbCommonBytes(register U32 val) { #if defined(LZ4_BIG_ENDIAN) #if ((defined(__GNUC__) && (GCC_VERSION >= 304)) || defined(__clang__)) && \ !defined(LZ4_FORCE_SW_BITCOUNT) return (__builtin_clz(val) >> 3); #else int r; if (!(val >> 16)) { r = 2; val >>= 8; } else { r = 0; val >>= 24; } r += (!val); return (r); #endif #else #if defined(__GNUC__) && (GCC_VERSION >= 304) && \ !defined(LZ4_FORCE_SW_BITCOUNT) return (__builtin_ctz(val) >> 3); #else static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, 3, 2, 2, 1, 3, 2, 0, 1, 3, 3, 1, 2, 2, 2, 2, 0, 3, 1, 2, 0, 1, 0, 1, 1 }; return DeBruijnBytePos[((U32) ((val & -(S32) val) * 0x077CB531U)) >> 27]; #endif #endif } #endif /* Compression functions */ static int LZ4_compressCtx(void *ctx, const char *source, char *dest, int isize, int osize) { struct refTables *srt = (struct refTables *)ctx; HTYPE *HashTable = (HTYPE *) (srt->hashTable); const BYTE *ip = (BYTE *) source; INITBASE(base); const BYTE *anchor = ip; const BYTE *const iend = ip + isize; const BYTE *const oend = (BYTE *) dest + osize; const BYTE *const mflimit = iend - MFLIMIT; #define matchlimit (iend - LASTLITERALS) BYTE *op = (BYTE *) dest; int len, length; const int skipStrength = SKIPSTRENGTH; U32 forwardH; /* Init */ if (isize < MINLENGTH) goto _last_literals; /* First Byte */ HashTable[LZ4_HASH_VALUE(ip)] = ip - base; ip++; forwardH = LZ4_HASH_VALUE(ip); /* Main Loop */ for (;;) { int findMatchAttempts = (1U << skipStrength) + 3; const BYTE *forwardIp = ip; const BYTE *ref; BYTE *token; /* Find a match */ do { U32 h = forwardH; int step = findMatchAttempts++ >> skipStrength; ip = forwardIp; forwardIp = ip + step; if (unlikely(forwardIp > mflimit)) { goto _last_literals; } forwardH = LZ4_HASH_VALUE(forwardIp); ref = base + HashTable[h]; HashTable[h] = ip - base; } while ((ref < ip - MAX_DISTANCE) || (A32(ref) != A32(ip))); /* Catch up */ while ((ip > anchor) && (ref > (BYTE *) source) && unlikely(ip[-1] == ref[-1])) { ip--; ref--; } /* Encode Literal length */ length = ip - anchor; token = op++; /* Check output limit */ if (unlikely(op + length + (2 + 1 + LASTLITERALS) + (length >> 8) > oend)) return (0); if (length >= (int)RUN_MASK) { *token = (RUN_MASK << ML_BITS); len = length - RUN_MASK; for (; len > 254; len -= 255) *op++ = 255; *op++ = (BYTE)len; } else *token = (length << ML_BITS); /* Copy Literals */ LZ4_BLINDCOPY(anchor, op, length); _next_match: /* Encode Offset */ LZ4_WRITE_LITTLEENDIAN_16(op, ip - ref); /* Start Counting */ ip += MINMATCH; ref += MINMATCH; /* MinMatch verified */ anchor = ip; while (likely(ip < matchlimit - (STEPSIZE - 1))) { UARCH diff = AARCH(ref) ^ AARCH(ip); if (!diff) { ip += STEPSIZE; ref += STEPSIZE; continue; } ip += LZ4_NbCommonBytes(diff); goto _endCount; } #if LZ4_ARCH64 if ((ip < (matchlimit - 3)) && (A32(ref) == A32(ip))) { ip += 4; ref += 4; } #endif if ((ip < (matchlimit - 1)) && (A16(ref) == A16(ip))) { ip += 2; ref += 2; } if ((ip < matchlimit) && (*ref == *ip)) ip++; _endCount: /* Encode MatchLength */ len = (ip - anchor); /* Check output limit */ if (unlikely(op + (1 + LASTLITERALS) + (len >> 8) > oend)) return (0); if (len >= (int)ML_MASK) { *token += ML_MASK; len -= ML_MASK; for (; len > 509; len -= 510) { *op++ = 255; *op++ = 255; } if (len > 254) { len -= 255; *op++ = 255; } *op++ = (BYTE)len; } else *token += len; /* Test end of chunk */ if (ip > mflimit) { anchor = ip; break; } /* Fill table */ HashTable[LZ4_HASH_VALUE(ip - 2)] = ip - 2 - base; /* Test next position */ ref = base + HashTable[LZ4_HASH_VALUE(ip)]; HashTable[LZ4_HASH_VALUE(ip)] = ip - base; if ((ref > ip - (MAX_DISTANCE + 1)) && (A32(ref) == A32(ip))) { token = op++; *token = 0; goto _next_match; } /* Prepare next loop */ anchor = ip++; forwardH = LZ4_HASH_VALUE(ip); } _last_literals: /* Encode Last Literals */ { int lastRun = iend - anchor; if (op + lastRun + 1 + ((lastRun + 255 - RUN_MASK) / 255) > oend) return (0); if (lastRun >= (int)RUN_MASK) { *op++ = (RUN_MASK << ML_BITS); lastRun -= RUN_MASK; for (; lastRun > 254; lastRun -= 255) { *op++ = 255; } *op++ = (BYTE)lastRun; } else *op++ = (lastRun << ML_BITS); (void) memcpy(op, anchor, iend - anchor); op += iend - anchor; } /* End */ return (int)(((char *)op) - dest); } /* Note : this function is valid only if isize < LZ4_64KLIMIT */ #define LZ4_64KLIMIT ((1 << 16) + (MFLIMIT - 1)) #define HASHLOG64K (HASH_LOG + 1) #define HASH64KTABLESIZE (1U << HASHLOG64K) #define LZ4_HASH64K_FUNCTION(i) (((i) * 2654435761U) >> ((MINMATCH*8) - \ HASHLOG64K)) #define LZ4_HASH64K_VALUE(p) LZ4_HASH64K_FUNCTION(A32(p)) static int LZ4_compress64kCtx(void *ctx, const char *source, char *dest, int isize, int osize) { struct refTables *srt = (struct refTables *)ctx; U16 *HashTable = (U16 *) (srt->hashTable); const BYTE *ip = (BYTE *) source; const BYTE *anchor = ip; const BYTE *const base = ip; const BYTE *const iend = ip + isize; const BYTE *const oend = (BYTE *) dest + osize; const BYTE *const mflimit = iend - MFLIMIT; #define matchlimit (iend - LASTLITERALS) BYTE *op = (BYTE *) dest; int len, length; const int skipStrength = SKIPSTRENGTH; U32 forwardH; /* Init */ if (isize < MINLENGTH) goto _last_literals; /* First Byte */ ip++; forwardH = LZ4_HASH64K_VALUE(ip); /* Main Loop */ for (;;) { int findMatchAttempts = (1U << skipStrength) + 3; const BYTE *forwardIp = ip; const BYTE *ref; BYTE *token; /* Find a match */ do { U32 h = forwardH; int step = findMatchAttempts++ >> skipStrength; ip = forwardIp; forwardIp = ip + step; if (forwardIp > mflimit) { goto _last_literals; } forwardH = LZ4_HASH64K_VALUE(forwardIp); ref = base + HashTable[h]; HashTable[h] = ip - base; } while (A32(ref) != A32(ip)); /* Catch up */ while ((ip > anchor) && (ref > (BYTE *) source) && (ip[-1] == ref[-1])) { ip--; ref--; } /* Encode Literal length */ length = ip - anchor; token = op++; /* Check output limit */ if (unlikely(op + length + (2 + 1 + LASTLITERALS) + (length >> 8) > oend)) return (0); if (length >= (int)RUN_MASK) { *token = (RUN_MASK << ML_BITS); len = length - RUN_MASK; for (; len > 254; len -= 255) *op++ = 255; *op++ = (BYTE)len; } else *token = (length << ML_BITS); /* Copy Literals */ LZ4_BLINDCOPY(anchor, op, length); _next_match: /* Encode Offset */ LZ4_WRITE_LITTLEENDIAN_16(op, ip - ref); /* Start Counting */ ip += MINMATCH; ref += MINMATCH; /* MinMatch verified */ anchor = ip; while (ip < matchlimit - (STEPSIZE - 1)) { UARCH diff = AARCH(ref) ^ AARCH(ip); if (!diff) { ip += STEPSIZE; ref += STEPSIZE; continue; } ip += LZ4_NbCommonBytes(diff); goto _endCount; } #if LZ4_ARCH64 if ((ip < (matchlimit - 3)) && (A32(ref) == A32(ip))) { ip += 4; ref += 4; } #endif if ((ip < (matchlimit - 1)) && (A16(ref) == A16(ip))) { ip += 2; ref += 2; } if ((ip < matchlimit) && (*ref == *ip)) ip++; _endCount: /* Encode MatchLength */ len = (ip - anchor); /* Check output limit */ if (unlikely(op + (1 + LASTLITERALS) + (len >> 8) > oend)) return (0); if (len >= (int)ML_MASK) { *token += ML_MASK; len -= ML_MASK; for (; len > 509; len -= 510) { *op++ = 255; *op++ = 255; } if (len > 254) { len -= 255; *op++ = 255; } *op++ = (BYTE)len; } else *token += len; /* Test end of chunk */ if (ip > mflimit) { anchor = ip; break; } /* Fill table */ HashTable[LZ4_HASH64K_VALUE(ip - 2)] = ip - 2 - base; /* Test next position */ ref = base + HashTable[LZ4_HASH64K_VALUE(ip)]; HashTable[LZ4_HASH64K_VALUE(ip)] = ip - base; if (A32(ref) == A32(ip)) { token = op++; *token = 0; goto _next_match; } /* Prepare next loop */ anchor = ip++; forwardH = LZ4_HASH64K_VALUE(ip); } _last_literals: /* Encode Last Literals */ { int lastRun = iend - anchor; if (op + lastRun + 1 + ((lastRun + 255 - RUN_MASK) / 255) > oend) return (0); if (lastRun >= (int)RUN_MASK) { *op++ = (RUN_MASK << ML_BITS); lastRun -= RUN_MASK; for (; lastRun > 254; lastRun -= 255) *op++ = 255; *op++ = (BYTE)lastRun; } else *op++ = (lastRun << ML_BITS); (void) memcpy(op, anchor, iend - anchor); op += iend - anchor; } /* End */ return (int)(((char *)op) - dest); } static int real_LZ4_compress(const char *source, char *dest, int isize, int osize) { void *ctx; int result; ASSERT(lz4_cache != NULL); ctx = kmem_cache_alloc(lz4_cache, KM_SLEEP); /* * out of kernel memory, gently fall through - this will disable * compression in zio_compress_data */ if (ctx == NULL) return (0); memset(ctx, 0, sizeof (struct refTables)); if (isize < LZ4_64KLIMIT) result = LZ4_compress64kCtx(ctx, source, dest, isize, osize); else result = LZ4_compressCtx(ctx, source, dest, isize, osize); kmem_cache_free(lz4_cache, ctx); return (result); } void lz4_init(void) { lz4_cache = kmem_cache_create("lz4_cache", sizeof (struct refTables), 0, NULL, NULL, NULL, NULL, NULL, KMC_RECLAIMABLE); } void lz4_fini(void) { if (lz4_cache) { kmem_cache_destroy(lz4_cache); lz4_cache = NULL; } } diff --git a/module/zfs/lzjb.c b/module/zfs/lzjb.c index b246693120ae..2db549b1626f 100644 --- a/module/zfs/lzjb.c +++ b/module/zfs/lzjb.c @@ -1,134 +1,137 @@ /* * 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. */ /* * We keep our own copy of this algorithm for 3 main reasons: * 1. If we didn't, anyone modifying common/os/compress.c would * directly break our on disk format * 2. Our version of lzjb does not have a number of checks that the * common/os version needs and uses * 3. We initialize the lempel to ensure deterministic results, * so that identical blocks can always be deduplicated. * In particular, we are adding the "feature" that compress() can * take a destination buffer size and returns the compressed length, or the * source length if compression would overflow the destination buffer. */ #include #include #define MATCH_BITS 6 #define MATCH_MIN 3 #define MATCH_MAX ((1 << MATCH_BITS) + (MATCH_MIN - 1)) #define OFFSET_MASK ((1 << (16 - MATCH_BITS)) - 1) #define LEMPEL_SIZE 1024 -size_t -zfs_lzjb_compress(void *s_start, void *d_start, size_t s_len, +static size_t +zfs_lzjb_compress_buf(void *s_start, void *d_start, size_t s_len, size_t d_len, int n) { (void) n; uchar_t *src = s_start; uchar_t *dst = d_start; uchar_t *cpy; uchar_t *copymap = NULL; int copymask = 1 << (NBBY - 1); int mlen, offset, hash; uint16_t *hp; uint16_t *lempel; lempel = kmem_zalloc(LEMPEL_SIZE * sizeof (uint16_t), KM_SLEEP); while (src < (uchar_t *)s_start + s_len) { if ((copymask <<= 1) == (1 << NBBY)) { if (dst >= (uchar_t *)d_start + d_len - 1 - 2 * NBBY) { kmem_free(lempel, LEMPEL_SIZE*sizeof (uint16_t)); return (s_len); } copymask = 1; copymap = dst; *dst++ = 0; } if (src > (uchar_t *)s_start + s_len - MATCH_MAX) { *dst++ = *src++; continue; } hash = (src[0] << 16) + (src[1] << 8) + src[2]; hash += hash >> 9; hash += hash >> 5; hp = &lempel[hash & (LEMPEL_SIZE - 1)]; offset = (intptr_t)(src - *hp) & OFFSET_MASK; *hp = (uint16_t)(uintptr_t)src; cpy = src - offset; if (cpy >= (uchar_t *)s_start && cpy != src && src[0] == cpy[0] && src[1] == cpy[1] && src[2] == cpy[2]) { *copymap |= copymask; for (mlen = MATCH_MIN; mlen < MATCH_MAX; mlen++) if (src[mlen] != cpy[mlen]) break; *dst++ = ((mlen - MATCH_MIN) << (NBBY - MATCH_BITS)) | (offset >> NBBY); *dst++ = (uchar_t)offset; src += mlen; } else { *dst++ = *src++; } } kmem_free(lempel, LEMPEL_SIZE * sizeof (uint16_t)); return (dst - (uchar_t *)d_start); } -int -zfs_lzjb_decompress(void *s_start, void *d_start, +static int +zfs_lzjb_decompress_buf(void *s_start, void *d_start, size_t s_len, size_t d_len, int n) { (void) s_len, (void) n; uchar_t *src = s_start; uchar_t *dst = d_start; uchar_t *d_end = (uchar_t *)d_start + d_len; uchar_t *cpy; uchar_t copymap = 0; int copymask = 1 << (NBBY - 1); while (dst < d_end) { if ((copymask <<= 1) == (1 << NBBY)) { copymask = 1; copymap = *src++; } if (copymap & copymask) { int mlen = (src[0] >> (NBBY - MATCH_BITS)) + MATCH_MIN; int offset = ((src[0] << NBBY) | src[1]) & OFFSET_MASK; src += 2; if ((cpy = dst - offset) < (uchar_t *)d_start) return (-1); while (--mlen >= 0 && dst < d_end) *dst++ = *cpy++; } else { *dst++ = *src++; } } return (0); } + +ZFS_COMPRESS_WRAP_DECL(zfs_lzjb_compress) +ZFS_DECOMPRESS_WRAP_DECL(zfs_lzjb_decompress) diff --git a/module/zfs/zio_compress.c b/module/zfs/zio_compress.c index 9553a9377c3d..118003bd295a 100644 --- a/module/zfs/zio_compress.c +++ b/module/zfs/zio_compress.c @@ -1,215 +1,216 @@ /* * 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 2009 Sun Microsystems, Inc. All rights reserved. * Use is subject to license terms. */ /* * Copyright (c) 2013 by Saso Kiselkov. All rights reserved. */ /* * Copyright (c) 2013, 2018 by Delphix. All rights reserved. - * Copyright (c) 2019, Klara Inc. + * Copyright (c) 2019, 2024, Klara, Inc. * Copyright (c) 2019, Allan Jude */ #include #include #include #include #include #include /* * If nonzero, every 1/X decompression attempts will fail, simulating * an undetected memory error. */ static unsigned long zio_decompress_fail_fraction = 0; /* * Compression vectors. * * NOTE: DO NOT CHANGE THE NAMES OF THESE COMPRESSION FUNCTIONS. * THEY ARE USED AS ZAP KEY NAMES BY FAST DEDUP AND THEREFORE * PART OF THE ON-DISK FORMAT. */ zio_compress_info_t zio_compress_table[ZIO_COMPRESS_FUNCTIONS] = { {"inherit", 0, NULL, NULL, NULL}, {"on", 0, NULL, NULL, NULL}, {"uncompressed", 0, NULL, NULL, NULL}, {"lzjb", 0, zfs_lzjb_compress, zfs_lzjb_decompress, NULL}, {"empty", 0, NULL, NULL, NULL}, {"gzip-1", 1, zfs_gzip_compress, zfs_gzip_decompress, NULL}, {"gzip-2", 2, zfs_gzip_compress, zfs_gzip_decompress, NULL}, {"gzip-3", 3, zfs_gzip_compress, zfs_gzip_decompress, NULL}, {"gzip-4", 4, zfs_gzip_compress, zfs_gzip_decompress, NULL}, {"gzip-5", 5, zfs_gzip_compress, zfs_gzip_decompress, NULL}, {"gzip-6", 6, zfs_gzip_compress, zfs_gzip_decompress, NULL}, {"gzip-7", 7, zfs_gzip_compress, zfs_gzip_decompress, NULL}, {"gzip-8", 8, zfs_gzip_compress, zfs_gzip_decompress, NULL}, {"gzip-9", 9, zfs_gzip_compress, zfs_gzip_decompress, NULL}, {"zle", 64, zfs_zle_compress, zfs_zle_decompress, NULL}, {"lz4", 0, zfs_lz4_compress, zfs_lz4_decompress, NULL}, {"zstd", ZIO_ZSTD_LEVEL_DEFAULT, zfs_zstd_compress, zfs_zstd_decompress, zfs_zstd_decompress_level}, }; uint8_t zio_complevel_select(spa_t *spa, enum zio_compress compress, uint8_t child, uint8_t parent) { (void) spa; uint8_t result; if (!ZIO_COMPRESS_HASLEVEL(compress)) return (0); result = child; if (result == ZIO_COMPLEVEL_INHERIT) result = parent; return (result); } enum zio_compress zio_compress_select(spa_t *spa, enum zio_compress child, enum zio_compress parent) { enum zio_compress result; ASSERT(child < ZIO_COMPRESS_FUNCTIONS); ASSERT(parent < ZIO_COMPRESS_FUNCTIONS); ASSERT(parent != ZIO_COMPRESS_INHERIT); result = child; if (result == ZIO_COMPRESS_INHERIT) result = parent; if (result == ZIO_COMPRESS_ON) { if (spa_feature_is_active(spa, SPA_FEATURE_LZ4_COMPRESS)) result = ZIO_COMPRESS_LZ4_ON_VALUE; else result = ZIO_COMPRESS_LEGACY_ON_VALUE; } return (result); } size_t zio_compress_data(enum zio_compress c, abd_t *src, void **dst, size_t s_len, uint8_t level) { size_t c_len, d_len; uint8_t complevel; zio_compress_info_t *ci = &zio_compress_table[c]; ASSERT3U(c, <, ZIO_COMPRESS_FUNCTIONS); ASSERT3U(ci->ci_compress, !=, NULL); ASSERT3U(s_len, >, 0); /* Compress at least 12.5% */ d_len = s_len - (s_len >> 3); complevel = ci->ci_level; if (c == ZIO_COMPRESS_ZSTD) { /* If we don't know the level, we can't compress it */ if (level == ZIO_COMPLEVEL_INHERIT) return (s_len); if (level == ZIO_COMPLEVEL_DEFAULT) complevel = ZIO_ZSTD_LEVEL_DEFAULT; else complevel = level; ASSERT3U(complevel, !=, ZIO_COMPLEVEL_INHERIT); } if (*dst == NULL) *dst = zio_buf_alloc(s_len); - /* No compression algorithms can read from ABDs directly */ - void *tmp = abd_borrow_buf_copy(src, s_len); - c_len = ci->ci_compress(tmp, *dst, s_len, d_len, complevel); - abd_return_buf(src, tmp, s_len); + abd_t dabd; + abd_get_from_buf_struct(&dabd, dst, d_len); + c_len = ci->ci_compress(src, &dabd, s_len, d_len, complevel); + abd_free(&dabd); if (c_len > d_len) return (s_len); ASSERT3U(c_len, <=, d_len); return (c_len); } int zio_decompress_data(enum zio_compress c, abd_t *src, void *dst, size_t s_len, size_t d_len, uint8_t *level) { zio_compress_info_t *ci = &zio_compress_table[c]; if ((uint_t)c >= ZIO_COMPRESS_FUNCTIONS || ci->ci_decompress == NULL) return (SET_ERROR(EINVAL)); - void *sbuf = abd_borrow_buf_copy(src, s_len); + abd_t dabd; + abd_get_from_buf_struct(&dabd, dst, d_len); int err; if (ci->ci_decompress_level != NULL && level != NULL) - err = ci->ci_decompress_level(sbuf, dst, s_len, d_len, level); + err = ci->ci_decompress_level(src, &dabd, s_len, d_len, level); else - err = ci->ci_decompress(sbuf, dst, s_len, d_len, ci->ci_level); + err = ci->ci_decompress(src, &dabd, s_len, d_len, ci->ci_level); - abd_return_buf(src, sbuf, s_len); + abd_free(&dabd); /* * Decompression shouldn't fail, because we've already verified * the checksum. However, for extra protection (e.g. against bitflips * in non-ECC RAM), we handle this error (and test it). */ if (zio_decompress_fail_fraction != 0 && random_in_range(zio_decompress_fail_fraction) == 0) err = SET_ERROR(EINVAL); return (err); } int zio_compress_to_feature(enum zio_compress comp) { switch (comp) { case ZIO_COMPRESS_ZSTD: return (SPA_FEATURE_ZSTD_COMPRESS); default: break; } return (SPA_FEATURE_NONE); } diff --git a/module/zfs/zle.c b/module/zfs/zle.c index 32b5fe18cec6..7810161966dc 100644 --- a/module/zfs/zle.c +++ b/module/zfs/zle.c @@ -1,93 +1,96 @@ /* * 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 2009 Sun Microsystems, Inc. All rights reserved. * Use is subject to license terms. */ /* * Zero-length encoding. This is a fast and simple algorithm to eliminate * runs of zeroes. Each chunk of compressed data begins with a length byte, b. * If b < n (where n is the compression parameter) then the next b + 1 bytes * are literal values. If b >= n then the next (256 - b + 1) bytes are zero. */ #include #include #include -size_t -zfs_zle_compress(void *s_start, void *d_start, size_t s_len, +static size_t +zfs_zle_compress_buf(void *s_start, void *d_start, size_t s_len, size_t d_len, int n) { uchar_t *src = s_start; uchar_t *dst = d_start; uchar_t *s_end = src + s_len; uchar_t *d_end = dst + d_len; while (src < s_end && dst < d_end - 1) { uchar_t *first = src; uchar_t *len = dst++; if (src[0] == 0) { uchar_t *last = src + (256 - n); while (src < MIN(last, s_end) && src[0] == 0) src++; *len = src - first - 1 + n; } else { uchar_t *last = src + n; if (d_end - dst < n) break; while (src < MIN(last, s_end) - 1 && (src[0] | src[1])) *dst++ = *src++; if (src[0]) *dst++ = *src++; *len = src - first - 1; } } return (src == s_end ? dst - (uchar_t *)d_start : s_len); } -int -zfs_zle_decompress(void *s_start, void *d_start, size_t s_len, +static int +zfs_zle_decompress_buf(void *s_start, void *d_start, size_t s_len, size_t d_len, int n) { uchar_t *src = s_start; uchar_t *dst = d_start; uchar_t *s_end = src + s_len; uchar_t *d_end = dst + d_len; while (src < s_end && dst < d_end) { int len = 1 + *src++; if (len <= n) { if (src + len > s_end || dst + len > d_end) return (-1); while (len-- != 0) *dst++ = *src++; } else { len -= n; if (dst + len > d_end) return (-1); while (len-- != 0) *dst++ = 0; } } return (dst == d_end ? 0 : -1); } + +ZFS_COMPRESS_WRAP_DECL(zfs_zle_compress) +ZFS_DECOMPRESS_WRAP_DECL(zfs_zle_decompress) diff --git a/module/zstd/zfs_zstd.c b/module/zstd/zfs_zstd.c index 34ab8fd8a424..8d1d53d234b4 100644 --- a/module/zstd/zfs_zstd.c +++ b/module/zstd/zfs_zstd.c @@ -1,911 +1,919 @@ /* * BSD 3-Clause New License (https://spdx.org/licenses/BSD-3-Clause.html) * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: * * 1. Redistributions of source code must retain the above copyright notice, * this list of conditions and the following disclaimer. * * 2. Redistributions in binary form must reproduce the above copyright notice, * this list of conditions and the following disclaimer in the documentation * and/or other materials provided with the distribution. * * 3. Neither the name of the copyright holder nor the names of its * contributors may be used to endorse or promote products derived from this * software without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE * POSSIBILITY OF SUCH DAMAGE. */ /* * Copyright (c) 2016-2018, Klara Inc. * Copyright (c) 2016-2018, Allan Jude * Copyright (c) 2018-2020, Sebastian Gottschall * Copyright (c) 2019-2020, Michael Niewöhner * Copyright (c) 2020, The FreeBSD Foundation [1] * * [1] Portions of this software were developed by Allan Jude * under sponsorship from the FreeBSD Foundation. */ #include #include #include #include #include #include #define ZSTD_STATIC_LINKING_ONLY #include "lib/zstd.h" #include "lib/common/zstd_errors.h" static uint_t zstd_earlyabort_pass = 1; static int zstd_cutoff_level = ZIO_ZSTD_LEVEL_3; static unsigned int zstd_abort_size = (128 * 1024); static kstat_t *zstd_ksp = NULL; typedef struct zstd_stats { kstat_named_t zstd_stat_alloc_fail; kstat_named_t zstd_stat_alloc_fallback; kstat_named_t zstd_stat_com_alloc_fail; kstat_named_t zstd_stat_dec_alloc_fail; kstat_named_t zstd_stat_com_inval; kstat_named_t zstd_stat_dec_inval; kstat_named_t zstd_stat_dec_header_inval; kstat_named_t zstd_stat_com_fail; kstat_named_t zstd_stat_dec_fail; /* * LZ4 first-pass early abort verdict */ kstat_named_t zstd_stat_lz4pass_allowed; kstat_named_t zstd_stat_lz4pass_rejected; /* * zstd-1 second-pass early abort verdict */ kstat_named_t zstd_stat_zstdpass_allowed; kstat_named_t zstd_stat_zstdpass_rejected; /* * We excluded this from early abort for some reason */ kstat_named_t zstd_stat_passignored; kstat_named_t zstd_stat_passignored_size; kstat_named_t zstd_stat_buffers; kstat_named_t zstd_stat_size; } zstd_stats_t; static zstd_stats_t zstd_stats = { { "alloc_fail", KSTAT_DATA_UINT64 }, { "alloc_fallback", KSTAT_DATA_UINT64 }, { "compress_alloc_fail", KSTAT_DATA_UINT64 }, { "decompress_alloc_fail", KSTAT_DATA_UINT64 }, { "compress_level_invalid", KSTAT_DATA_UINT64 }, { "decompress_level_invalid", KSTAT_DATA_UINT64 }, { "decompress_header_invalid", KSTAT_DATA_UINT64 }, { "compress_failed", KSTAT_DATA_UINT64 }, { "decompress_failed", KSTAT_DATA_UINT64 }, { "lz4pass_allowed", KSTAT_DATA_UINT64 }, { "lz4pass_rejected", KSTAT_DATA_UINT64 }, { "zstdpass_allowed", KSTAT_DATA_UINT64 }, { "zstdpass_rejected", KSTAT_DATA_UINT64 }, { "passignored", KSTAT_DATA_UINT64 }, { "passignored_size", KSTAT_DATA_UINT64 }, { "buffers", KSTAT_DATA_UINT64 }, { "size", KSTAT_DATA_UINT64 }, }; #ifdef _KERNEL static int kstat_zstd_update(kstat_t *ksp, int rw) { ASSERT(ksp != NULL); if (rw == KSTAT_WRITE && ksp == zstd_ksp) { ZSTDSTAT_ZERO(zstd_stat_alloc_fail); ZSTDSTAT_ZERO(zstd_stat_alloc_fallback); ZSTDSTAT_ZERO(zstd_stat_com_alloc_fail); ZSTDSTAT_ZERO(zstd_stat_dec_alloc_fail); ZSTDSTAT_ZERO(zstd_stat_com_inval); ZSTDSTAT_ZERO(zstd_stat_dec_inval); ZSTDSTAT_ZERO(zstd_stat_dec_header_inval); ZSTDSTAT_ZERO(zstd_stat_com_fail); ZSTDSTAT_ZERO(zstd_stat_dec_fail); ZSTDSTAT_ZERO(zstd_stat_lz4pass_allowed); ZSTDSTAT_ZERO(zstd_stat_lz4pass_rejected); ZSTDSTAT_ZERO(zstd_stat_zstdpass_allowed); ZSTDSTAT_ZERO(zstd_stat_zstdpass_rejected); ZSTDSTAT_ZERO(zstd_stat_passignored); ZSTDSTAT_ZERO(zstd_stat_passignored_size); } return (0); } #endif /* Enums describing the allocator type specified by kmem_type in zstd_kmem */ enum zstd_kmem_type { ZSTD_KMEM_UNKNOWN = 0, /* Allocation type using kmem_vmalloc */ ZSTD_KMEM_DEFAULT, /* Pool based allocation using mempool_alloc */ ZSTD_KMEM_POOL, /* Reserved fallback memory for decompression only */ ZSTD_KMEM_DCTX, ZSTD_KMEM_COUNT, }; /* Structure for pooled memory objects */ struct zstd_pool { void *mem; size_t size; kmutex_t barrier; hrtime_t timeout; }; /* Global structure for handling memory allocations */ struct zstd_kmem { enum zstd_kmem_type kmem_type; size_t kmem_size; struct zstd_pool *pool; }; /* Fallback memory structure used for decompression only if memory runs out */ struct zstd_fallback_mem { size_t mem_size; void *mem; kmutex_t barrier; }; struct zstd_levelmap { int16_t zstd_level; enum zio_zstd_levels level; }; /* * ZSTD memory handlers * * For decompression we use a different handler which also provides fallback * memory allocation in case memory runs out. * * The ZSTD handlers were split up for the most simplified implementation. */ static void *zstd_alloc(void *opaque, size_t size); static void *zstd_dctx_alloc(void *opaque, size_t size); static void zstd_free(void *opaque, void *ptr); /* Compression memory handler */ static const ZSTD_customMem zstd_malloc = { zstd_alloc, zstd_free, NULL, }; /* Decompression memory handler */ static const ZSTD_customMem zstd_dctx_malloc = { zstd_dctx_alloc, zstd_free, NULL, }; /* Level map for converting ZFS internal levels to ZSTD levels and vice versa */ static struct zstd_levelmap zstd_levels[] = { {ZIO_ZSTD_LEVEL_1, ZIO_ZSTD_LEVEL_1}, {ZIO_ZSTD_LEVEL_2, ZIO_ZSTD_LEVEL_2}, {ZIO_ZSTD_LEVEL_3, ZIO_ZSTD_LEVEL_3}, {ZIO_ZSTD_LEVEL_4, ZIO_ZSTD_LEVEL_4}, {ZIO_ZSTD_LEVEL_5, ZIO_ZSTD_LEVEL_5}, {ZIO_ZSTD_LEVEL_6, ZIO_ZSTD_LEVEL_6}, {ZIO_ZSTD_LEVEL_7, ZIO_ZSTD_LEVEL_7}, {ZIO_ZSTD_LEVEL_8, ZIO_ZSTD_LEVEL_8}, {ZIO_ZSTD_LEVEL_9, ZIO_ZSTD_LEVEL_9}, {ZIO_ZSTD_LEVEL_10, ZIO_ZSTD_LEVEL_10}, {ZIO_ZSTD_LEVEL_11, ZIO_ZSTD_LEVEL_11}, {ZIO_ZSTD_LEVEL_12, ZIO_ZSTD_LEVEL_12}, {ZIO_ZSTD_LEVEL_13, ZIO_ZSTD_LEVEL_13}, {ZIO_ZSTD_LEVEL_14, ZIO_ZSTD_LEVEL_14}, {ZIO_ZSTD_LEVEL_15, ZIO_ZSTD_LEVEL_15}, {ZIO_ZSTD_LEVEL_16, ZIO_ZSTD_LEVEL_16}, {ZIO_ZSTD_LEVEL_17, ZIO_ZSTD_LEVEL_17}, {ZIO_ZSTD_LEVEL_18, ZIO_ZSTD_LEVEL_18}, {ZIO_ZSTD_LEVEL_19, ZIO_ZSTD_LEVEL_19}, {-1, ZIO_ZSTD_LEVEL_FAST_1}, {-2, ZIO_ZSTD_LEVEL_FAST_2}, {-3, ZIO_ZSTD_LEVEL_FAST_3}, {-4, ZIO_ZSTD_LEVEL_FAST_4}, {-5, ZIO_ZSTD_LEVEL_FAST_5}, {-6, ZIO_ZSTD_LEVEL_FAST_6}, {-7, ZIO_ZSTD_LEVEL_FAST_7}, {-8, ZIO_ZSTD_LEVEL_FAST_8}, {-9, ZIO_ZSTD_LEVEL_FAST_9}, {-10, ZIO_ZSTD_LEVEL_FAST_10}, {-20, ZIO_ZSTD_LEVEL_FAST_20}, {-30, ZIO_ZSTD_LEVEL_FAST_30}, {-40, ZIO_ZSTD_LEVEL_FAST_40}, {-50, ZIO_ZSTD_LEVEL_FAST_50}, {-60, ZIO_ZSTD_LEVEL_FAST_60}, {-70, ZIO_ZSTD_LEVEL_FAST_70}, {-80, ZIO_ZSTD_LEVEL_FAST_80}, {-90, ZIO_ZSTD_LEVEL_FAST_90}, {-100, ZIO_ZSTD_LEVEL_FAST_100}, {-500, ZIO_ZSTD_LEVEL_FAST_500}, {-1000, ZIO_ZSTD_LEVEL_FAST_1000}, }; /* * This variable represents the maximum count of the pool based on the number * of CPUs plus some buffer. We default to cpu count * 4, see init_zstd. */ static int pool_count = 16; #define ZSTD_POOL_MAX pool_count #define ZSTD_POOL_TIMEOUT 60 * 2 static struct zstd_fallback_mem zstd_dctx_fallback; static struct zstd_pool *zstd_mempool_cctx; static struct zstd_pool *zstd_mempool_dctx; /* * The library zstd code expects these if ADDRESS_SANITIZER gets defined, * and while ASAN does this, KASAN defines that and does not. So to avoid * changing the external code, we do this. */ #if defined(ZFS_ASAN_ENABLED) #define ADDRESS_SANITIZER 1 #endif #if defined(_KERNEL) && defined(ADDRESS_SANITIZER) void __asan_unpoison_memory_region(void const volatile *addr, size_t size); void __asan_poison_memory_region(void const volatile *addr, size_t size); void __asan_unpoison_memory_region(void const volatile *addr, size_t size) {}; void __asan_poison_memory_region(void const volatile *addr, size_t size) {}; #endif static void zstd_mempool_reap(struct zstd_pool *zstd_mempool) { struct zstd_pool *pool; if (!zstd_mempool || !ZSTDSTAT(zstd_stat_buffers)) { return; } /* free obsolete slots */ for (int i = 0; i < ZSTD_POOL_MAX; i++) { pool = &zstd_mempool[i]; if (pool->mem && mutex_tryenter(&pool->barrier)) { /* Free memory if unused object older than 2 minutes */ if (pool->mem && gethrestime_sec() > pool->timeout) { vmem_free(pool->mem, pool->size); ZSTDSTAT_SUB(zstd_stat_buffers, 1); ZSTDSTAT_SUB(zstd_stat_size, pool->size); pool->mem = NULL; pool->size = 0; pool->timeout = 0; } mutex_exit(&pool->barrier); } } } /* * Try to get a cached allocated buffer from memory pool or allocate a new one * if necessary. If a object is older than 2 minutes and does not fit the * requested size, it will be released and a new cached entry will be allocated. * If other pooled objects are detected without being used for 2 minutes, they * will be released, too. * * The concept is that high frequency memory allocations of bigger objects are * expensive. So if a lot of work is going on, allocations will be kept for a * while and can be reused in that time frame. * * The scheduled release will be updated every time a object is reused. */ static void * zstd_mempool_alloc(struct zstd_pool *zstd_mempool, size_t size) { struct zstd_pool *pool; struct zstd_kmem *mem = NULL; if (!zstd_mempool) { return (NULL); } /* Seek for preallocated memory slot and free obsolete slots */ for (int i = 0; i < ZSTD_POOL_MAX; i++) { pool = &zstd_mempool[i]; /* * This lock is simply a marker for a pool object being in use. * If it's already hold, it will be skipped. * * We need to create it before checking it to avoid race * conditions caused by running in a threaded context. * * The lock is later released by zstd_mempool_free. */ if (mutex_tryenter(&pool->barrier)) { /* * Check if objects fits the size, if so we take it and * update the timestamp. */ if (pool->mem && size <= pool->size) { pool->timeout = gethrestime_sec() + ZSTD_POOL_TIMEOUT; mem = pool->mem; return (mem); } mutex_exit(&pool->barrier); } } /* * If no preallocated slot was found, try to fill in a new one. * * We run a similar algorithm twice here to avoid pool fragmentation. * The first one may generate holes in the list if objects get released. * We always make sure that these holes get filled instead of adding new * allocations constantly at the end. */ for (int i = 0; i < ZSTD_POOL_MAX; i++) { pool = &zstd_mempool[i]; if (mutex_tryenter(&pool->barrier)) { /* Object is free, try to allocate new one */ if (!pool->mem) { mem = vmem_alloc(size, KM_SLEEP); if (mem) { ZSTDSTAT_ADD(zstd_stat_buffers, 1); ZSTDSTAT_ADD(zstd_stat_size, size); pool->mem = mem; pool->size = size; /* Keep track for later release */ mem->pool = pool; mem->kmem_type = ZSTD_KMEM_POOL; mem->kmem_size = size; } } if (size <= pool->size) { /* Update timestamp */ pool->timeout = gethrestime_sec() + ZSTD_POOL_TIMEOUT; return (pool->mem); } mutex_exit(&pool->barrier); } } /* * If the pool is full or the allocation failed, try lazy allocation * instead. */ if (!mem) { mem = vmem_alloc(size, KM_NOSLEEP); if (mem) { mem->pool = NULL; mem->kmem_type = ZSTD_KMEM_DEFAULT; mem->kmem_size = size; } } return (mem); } /* Mark object as released by releasing the barrier mutex */ static void zstd_mempool_free(struct zstd_kmem *z) { mutex_exit(&z->pool->barrier); } /* Convert ZFS internal enum to ZSTD level */ static int zstd_enum_to_level(enum zio_zstd_levels level, int16_t *zstd_level) { if (level > 0 && level <= ZIO_ZSTD_LEVEL_19) { *zstd_level = zstd_levels[level - 1].zstd_level; return (0); } if (level >= ZIO_ZSTD_LEVEL_FAST_1 && level <= ZIO_ZSTD_LEVEL_FAST_1000) { *zstd_level = zstd_levels[level - ZIO_ZSTD_LEVEL_FAST_1 + ZIO_ZSTD_LEVEL_19].zstd_level; return (0); } /* Invalid/unknown zfs compression enum - this should never happen. */ return (1); } /* Compress block using zstd */ static size_t zfs_zstd_compress_impl(void *s_start, void *d_start, size_t s_len, size_t d_len, int level) { size_t c_len; int16_t zstd_level; zfs_zstdhdr_t *hdr; ZSTD_CCtx *cctx; hdr = (zfs_zstdhdr_t *)d_start; /* Skip compression if the specified level is invalid */ if (zstd_enum_to_level(level, &zstd_level)) { ZSTDSTAT_BUMP(zstd_stat_com_inval); return (s_len); } ASSERT3U(d_len, >=, sizeof (*hdr)); ASSERT3U(d_len, <=, s_len); ASSERT3U(zstd_level, !=, 0); cctx = ZSTD_createCCtx_advanced(zstd_malloc); /* * Out of kernel memory, gently fall through - this will disable * compression in zio_compress_data */ if (!cctx) { ZSTDSTAT_BUMP(zstd_stat_com_alloc_fail); return (s_len); } /* Set the compression level */ ZSTD_CCtx_setParameter(cctx, ZSTD_c_compressionLevel, zstd_level); /* Use the "magicless" zstd header which saves us 4 header bytes */ ZSTD_CCtx_setParameter(cctx, ZSTD_c_format, ZSTD_f_zstd1_magicless); /* * Disable redundant checksum calculation and content size storage since * this is already done by ZFS itself. */ ZSTD_CCtx_setParameter(cctx, ZSTD_c_checksumFlag, 0); ZSTD_CCtx_setParameter(cctx, ZSTD_c_contentSizeFlag, 0); c_len = ZSTD_compress2(cctx, hdr->data, d_len - sizeof (*hdr), s_start, s_len); ZSTD_freeCCtx(cctx); /* Error in the compression routine, disable compression. */ if (ZSTD_isError(c_len)) { /* * If we are aborting the compression because the saves are * too small, that is not a failure. Everything else is a * failure, so increment the compression failure counter. */ int err = ZSTD_getErrorCode(c_len); if (err != ZSTD_error_dstSize_tooSmall) { ZSTDSTAT_BUMP(zstd_stat_com_fail); dprintf("Error: %s", ZSTD_getErrorString(err)); } return (s_len); } /* * Encode the compressed buffer size at the start. We'll need this in * decompression to counter the effects of padding which might be added * to the compressed buffer and which, if unhandled, would confuse the * hell out of our decompression function. */ hdr->c_len = BE_32(c_len); /* * Check version for overflow. * The limit of 24 bits must not be exceeded. This allows a maximum * version 1677.72.15 which we don't expect to be ever reached. */ ASSERT3U(ZSTD_VERSION_NUMBER, <=, 0xFFFFFF); /* * Encode the compression level as well. We may need to know the * original compression level if compressed_arc is disabled, to match * the compression settings to write this block to the L2ARC. * * Encode the actual level, so if the enum changes in the future, we * will be compatible. * * The upper 24 bits store the ZSTD version to be able to provide * future compatibility, since new versions might enhance the * compression algorithm in a way, where the compressed data will * change. * * As soon as such incompatibility occurs, handling code needs to be * added, differentiating between the versions. */ zfs_set_hdrversion(hdr, ZSTD_VERSION_NUMBER); zfs_set_hdrlevel(hdr, level); hdr->raw_version_level = BE_32(hdr->raw_version_level); return (c_len + sizeof (*hdr)); } -size_t -zfs_zstd_compress(void *s_start, void *d_start, size_t s_len, size_t d_len, +static size_t +zfs_zstd_compress_buf(void *s_start, void *d_start, size_t s_len, size_t d_len, int level) { int16_t zstd_level; if (zstd_enum_to_level(level, &zstd_level)) { ZSTDSTAT_BUMP(zstd_stat_com_inval); return (s_len); } /* * A zstd early abort heuristic. * * - Zeroth, if this is <= zstd-3, or < zstd_abort_size (currently * 128k), don't try any of this, just go. * (because experimentally that was a reasonable cutoff for a perf win * with tiny ratio change) * - First, we try LZ4 compression, and if it doesn't early abort, we * jump directly to whatever compression level we intended to try. * - Second, we try zstd-1 - if that errors out (usually, but not * exclusively, if it would overflow), we give up early. * * If it works, instead we go on and compress anyway. * * Why two passes? LZ4 alone gets you a lot of the way, but on highly * compressible data, it was losing up to 8.5% of the compressed * savings versus no early abort, and all the zstd-fast levels are * worse indications on their own than LZ4, and don't improve the LZ4 * pass noticably if stacked like this. */ size_t actual_abort_size = zstd_abort_size; if (zstd_earlyabort_pass > 0 && zstd_level >= zstd_cutoff_level && s_len >= actual_abort_size) { int pass_len = 1; - pass_len = zfs_lz4_compress(s_start, d_start, s_len, d_len, 0); + abd_t sabd; + abd_get_from_buf_struct(&sabd, s_start, s_len); + pass_len = zfs_lz4_compress(&sabd, d_start, s_len, d_len, 0); + abd_free(&sabd); if (pass_len < d_len) { ZSTDSTAT_BUMP(zstd_stat_lz4pass_allowed); goto keep_trying; } ZSTDSTAT_BUMP(zstd_stat_lz4pass_rejected); pass_len = zfs_zstd_compress_impl(s_start, d_start, s_len, d_len, ZIO_ZSTD_LEVEL_1); if (pass_len == s_len || pass_len <= 0 || pass_len > d_len) { ZSTDSTAT_BUMP(zstd_stat_zstdpass_rejected); return (s_len); } ZSTDSTAT_BUMP(zstd_stat_zstdpass_allowed); } else { ZSTDSTAT_BUMP(zstd_stat_passignored); if (s_len < actual_abort_size) { ZSTDSTAT_BUMP(zstd_stat_passignored_size); } } keep_trying: return (zfs_zstd_compress_impl(s_start, d_start, s_len, d_len, level)); } /* Decompress block using zstd and return its stored level */ -int -zfs_zstd_decompress_level(void *s_start, void *d_start, size_t s_len, +static int +zfs_zstd_decompress_level_buf(void *s_start, void *d_start, size_t s_len, size_t d_len, uint8_t *level) { ZSTD_DCtx *dctx; size_t result; int16_t zstd_level; uint32_t c_len; const zfs_zstdhdr_t *hdr; zfs_zstdhdr_t hdr_copy; hdr = (const zfs_zstdhdr_t *)s_start; c_len = BE_32(hdr->c_len); /* * Make a copy instead of directly converting the header, since we must * not modify the original data that may be used again later. */ hdr_copy.raw_version_level = BE_32(hdr->raw_version_level); uint8_t curlevel = zfs_get_hdrlevel(&hdr_copy); /* * NOTE: We ignore the ZSTD version for now. As soon as any * incompatibility occurs, it has to be handled accordingly. * The version can be accessed via `hdr_copy.version`. */ /* * Convert and check the level * An invalid level is a strong indicator for data corruption! In such * case return an error so the upper layers can try to fix it. */ if (zstd_enum_to_level(curlevel, &zstd_level)) { ZSTDSTAT_BUMP(zstd_stat_dec_inval); return (1); } ASSERT3U(d_len, >=, s_len); ASSERT3U(curlevel, !=, ZIO_COMPLEVEL_INHERIT); /* Invalid compressed buffer size encoded at start */ if (c_len + sizeof (*hdr) > s_len) { ZSTDSTAT_BUMP(zstd_stat_dec_header_inval); return (1); } dctx = ZSTD_createDCtx_advanced(zstd_dctx_malloc); if (!dctx) { ZSTDSTAT_BUMP(zstd_stat_dec_alloc_fail); return (1); } /* Set header type to "magicless" */ ZSTD_DCtx_setParameter(dctx, ZSTD_d_format, ZSTD_f_zstd1_magicless); /* Decompress the data and release the context */ result = ZSTD_decompressDCtx(dctx, d_start, d_len, hdr->data, c_len); ZSTD_freeDCtx(dctx); /* * Returns 0 on success (decompression function returned non-negative) * and non-zero on failure (decompression function returned negative. */ if (ZSTD_isError(result)) { ZSTDSTAT_BUMP(zstd_stat_dec_fail); return (1); } if (level) { *level = curlevel; } return (0); } /* Decompress datablock using zstd */ -int -zfs_zstd_decompress(void *s_start, void *d_start, size_t s_len, size_t d_len, - int level __maybe_unused) +static int +zfs_zstd_decompress_buf(void *s_start, void *d_start, size_t s_len, + size_t d_len, int level __maybe_unused) { - return (zfs_zstd_decompress_level(s_start, d_start, s_len, d_len, + return (zfs_zstd_decompress_level_buf(s_start, d_start, s_len, d_len, NULL)); } +ZFS_COMPRESS_WRAP_DECL(zfs_zstd_compress) +ZFS_DECOMPRESS_WRAP_DECL(zfs_zstd_decompress) +ZFS_DECOMPRESS_LEVEL_WRAP_DECL(zfs_zstd_decompress_level) + + /* Allocator for zstd compression context using mempool_allocator */ static void * zstd_alloc(void *opaque __maybe_unused, size_t size) { size_t nbytes = sizeof (struct zstd_kmem) + size; struct zstd_kmem *z = NULL; z = (struct zstd_kmem *)zstd_mempool_alloc(zstd_mempool_cctx, nbytes); if (!z) { ZSTDSTAT_BUMP(zstd_stat_alloc_fail); return (NULL); } return ((void*)z + (sizeof (struct zstd_kmem))); } /* * Allocator for zstd decompression context using mempool_allocator with * fallback to reserved memory if allocation fails */ static void * zstd_dctx_alloc(void *opaque __maybe_unused, size_t size) { size_t nbytes = sizeof (struct zstd_kmem) + size; struct zstd_kmem *z = NULL; enum zstd_kmem_type type = ZSTD_KMEM_DEFAULT; z = (struct zstd_kmem *)zstd_mempool_alloc(zstd_mempool_dctx, nbytes); if (!z) { /* Try harder, decompression shall not fail */ z = vmem_alloc(nbytes, KM_SLEEP); if (z) { z->pool = NULL; } ZSTDSTAT_BUMP(zstd_stat_alloc_fail); } else { return ((void*)z + (sizeof (struct zstd_kmem))); } /* Fallback if everything fails */ if (!z) { /* * Barrier since we only can handle it in a single thread. All * other following threads need to wait here until decompression * is completed. zstd_free will release this barrier later. */ mutex_enter(&zstd_dctx_fallback.barrier); z = zstd_dctx_fallback.mem; type = ZSTD_KMEM_DCTX; ZSTDSTAT_BUMP(zstd_stat_alloc_fallback); } /* Allocation should always be successful */ if (!z) { return (NULL); } z->kmem_type = type; z->kmem_size = nbytes; return ((void*)z + (sizeof (struct zstd_kmem))); } /* Free allocated memory by its specific type */ static void zstd_free(void *opaque __maybe_unused, void *ptr) { struct zstd_kmem *z = (ptr - sizeof (struct zstd_kmem)); enum zstd_kmem_type type; ASSERT3U(z->kmem_type, <, ZSTD_KMEM_COUNT); ASSERT3U(z->kmem_type, >, ZSTD_KMEM_UNKNOWN); type = z->kmem_type; switch (type) { case ZSTD_KMEM_DEFAULT: vmem_free(z, z->kmem_size); break; case ZSTD_KMEM_POOL: zstd_mempool_free(z); break; case ZSTD_KMEM_DCTX: mutex_exit(&zstd_dctx_fallback.barrier); break; default: break; } } /* Allocate fallback memory to ensure safe decompression */ static void __init create_fallback_mem(struct zstd_fallback_mem *mem, size_t size) { mem->mem_size = size; mem->mem = vmem_zalloc(mem->mem_size, KM_SLEEP); mutex_init(&mem->barrier, NULL, MUTEX_DEFAULT, NULL); } /* Initialize memory pool barrier mutexes */ static void __init zstd_mempool_init(void) { zstd_mempool_cctx = kmem_zalloc(ZSTD_POOL_MAX * sizeof (struct zstd_pool), KM_SLEEP); zstd_mempool_dctx = kmem_zalloc(ZSTD_POOL_MAX * sizeof (struct zstd_pool), KM_SLEEP); for (int i = 0; i < ZSTD_POOL_MAX; i++) { mutex_init(&zstd_mempool_cctx[i].barrier, NULL, MUTEX_DEFAULT, NULL); mutex_init(&zstd_mempool_dctx[i].barrier, NULL, MUTEX_DEFAULT, NULL); } } /* Initialize zstd-related memory handling */ static int __init zstd_meminit(void) { zstd_mempool_init(); /* * Estimate the size of the fallback decompression context. * The expected size on x64 with current ZSTD should be about 160 KB. */ create_fallback_mem(&zstd_dctx_fallback, P2ROUNDUP(ZSTD_estimateDCtxSize() + sizeof (struct zstd_kmem), PAGESIZE)); return (0); } /* Release object from pool and free memory */ static void release_pool(struct zstd_pool *pool) { mutex_destroy(&pool->barrier); vmem_free(pool->mem, pool->size); pool->mem = NULL; pool->size = 0; } /* Release memory pool objects */ static void zstd_mempool_deinit(void) { for (int i = 0; i < ZSTD_POOL_MAX; i++) { release_pool(&zstd_mempool_cctx[i]); release_pool(&zstd_mempool_dctx[i]); } kmem_free(zstd_mempool_dctx, ZSTD_POOL_MAX * sizeof (struct zstd_pool)); kmem_free(zstd_mempool_cctx, ZSTD_POOL_MAX * sizeof (struct zstd_pool)); zstd_mempool_dctx = NULL; zstd_mempool_cctx = NULL; } /* release unused memory from pool */ void zfs_zstd_cache_reap_now(void) { /* * Short-circuit if there are no buffers to begin with. */ if (ZSTDSTAT(zstd_stat_buffers) == 0) return; /* * calling alloc with zero size seeks * and releases old unused objects */ zstd_mempool_reap(zstd_mempool_cctx); zstd_mempool_reap(zstd_mempool_dctx); } extern int __init zstd_init(void) { /* Set pool size by using maximum sane thread count * 4 */ pool_count = (boot_ncpus * 4); zstd_meminit(); /* Initialize kstat */ zstd_ksp = kstat_create("zfs", 0, "zstd", "misc", KSTAT_TYPE_NAMED, sizeof (zstd_stats) / sizeof (kstat_named_t), KSTAT_FLAG_VIRTUAL); if (zstd_ksp != NULL) { zstd_ksp->ks_data = &zstd_stats; kstat_install(zstd_ksp); #ifdef _KERNEL zstd_ksp->ks_update = kstat_zstd_update; #endif } return (0); } extern void zstd_fini(void) { /* Deinitialize kstat */ if (zstd_ksp != NULL) { kstat_delete(zstd_ksp); zstd_ksp = NULL; } /* Release fallback memory */ vmem_free(zstd_dctx_fallback.mem, zstd_dctx_fallback.mem_size); mutex_destroy(&zstd_dctx_fallback.barrier); /* Deinit memory pool */ zstd_mempool_deinit(); } #if defined(_KERNEL) #ifdef __FreeBSD__ module_init(zstd_init); module_exit(zstd_fini); #endif ZFS_MODULE_PARAM(zfs, zstd_, earlyabort_pass, UINT, ZMOD_RW, "Enable early abort attempts when using zstd"); ZFS_MODULE_PARAM(zfs, zstd_, abort_size, UINT, ZMOD_RW, "Minimal size of block to attempt early abort"); #endif