diff options
Diffstat (limited to 'fs/f2fs/super.c')
-rw-r--r-- | fs/f2fs/super.c | 1699 |
1 files changed, 1275 insertions, 424 deletions
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 65a7a432dfee..3834ead04620 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -8,9 +8,10 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/fs.h> +#include <linux/fs_context.h> +#include <linux/sched/mm.h> #include <linux/statfs.h> #include <linux/buffer_head.h> -#include <linux/backing-dev.h> #include <linux/kthread.h> #include <linux/parser.h> #include <linux/mount.h> @@ -24,13 +25,16 @@ #include <linux/sysfs.h> #include <linux/quota.h> #include <linux/unicode.h> +#include <linux/part_stat.h> +#include <linux/zstd.h> +#include <linux/lz4.h> #include "f2fs.h" #include "node.h" #include "segment.h" #include "xattr.h" #include "gc.h" -#include "trace.h" +#include "iostat.h" #define CREATE_TRACE_POINTS #include <trace/events/f2fs.h> @@ -44,7 +48,6 @@ const char *f2fs_fault_name[FAULT_MAX] = { [FAULT_KVMALLOC] = "kvmalloc", [FAULT_PAGE_ALLOC] = "page alloc", [FAULT_PAGE_GET] = "page get", - [FAULT_ALLOC_BIO] = "alloc bio", [FAULT_ALLOC_NID] = "alloc nid", [FAULT_ORPHAN] = "orphan", [FAULT_BLOCK] = "no more block", @@ -55,6 +58,9 @@ const char *f2fs_fault_name[FAULT_MAX] = { [FAULT_CHECKPOINT] = "checkpoint error", [FAULT_DISCARD] = "discard error", [FAULT_WRITE_IO] = "write IO error", + [FAULT_SLAB_ALLOC] = "slab alloc", + [FAULT_DQUOT_INIT] = "dquot initialize", + [FAULT_LOCK_OP] = "lock_op", }; void f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned int rate, @@ -133,17 +139,28 @@ enum { Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, - Opt_whint, Opt_alloc, Opt_fsync, Opt_test_dummy_encryption, + Opt_inlinecrypt, Opt_checkpoint_disable, Opt_checkpoint_disable_cap, Opt_checkpoint_disable_cap_perc, Opt_checkpoint_enable, + Opt_checkpoint_merge, + Opt_nocheckpoint_merge, Opt_compress_algorithm, Opt_compress_log_size, Opt_compress_extension, + Opt_nocompress_extension, + Opt_compress_chksum, + Opt_compress_mode, + Opt_compress_cache, + Opt_atgc, + Opt_gc_merge, + Opt_nogc_merge, + Opt_discard_unit, + Opt_memory_mode, Opt_err, }; @@ -198,17 +215,29 @@ static match_table_t f2fs_tokens = { {Opt_jqfmt_vfsold, "jqfmt=vfsold"}, {Opt_jqfmt_vfsv0, "jqfmt=vfsv0"}, {Opt_jqfmt_vfsv1, "jqfmt=vfsv1"}, - {Opt_whint, "whint_mode=%s"}, {Opt_alloc, "alloc_mode=%s"}, {Opt_fsync, "fsync_mode=%s"}, + {Opt_test_dummy_encryption, "test_dummy_encryption=%s"}, {Opt_test_dummy_encryption, "test_dummy_encryption"}, + {Opt_inlinecrypt, "inlinecrypt"}, {Opt_checkpoint_disable, "checkpoint=disable"}, {Opt_checkpoint_disable_cap, "checkpoint=disable:%u"}, {Opt_checkpoint_disable_cap_perc, "checkpoint=disable:%u%%"}, {Opt_checkpoint_enable, "checkpoint=enable"}, + {Opt_checkpoint_merge, "checkpoint_merge"}, + {Opt_nocheckpoint_merge, "nocheckpoint_merge"}, {Opt_compress_algorithm, "compress_algorithm=%s"}, {Opt_compress_log_size, "compress_log_size=%u"}, {Opt_compress_extension, "compress_extension=%s"}, + {Opt_nocompress_extension, "nocompress_extension=%s"}, + {Opt_compress_chksum, "compress_chksum"}, + {Opt_compress_mode, "compress_mode=%s"}, + {Opt_compress_cache, "compress_cache"}, + {Opt_atgc, "atgc"}, + {Opt_gc_merge, "gc_merge"}, + {Opt_nogc_merge, "nogc_merge"}, + {Opt_discard_unit, "discard_unit=%s"}, + {Opt_memory_mode, "memory=%s"}, {Opt_err, NULL}, }; @@ -229,42 +258,53 @@ void f2fs_printk(struct f2fs_sb_info *sbi, const char *fmt, ...) va_end(args); } -#ifdef CONFIG_UNICODE +#if IS_ENABLED(CONFIG_UNICODE) static const struct f2fs_sb_encodings { __u16 magic; char *name; - char *version; + unsigned int version; } f2fs_sb_encoding_map[] = { - {F2FS_ENC_UTF8_12_1, "utf8", "12.1.0"}, + {F2FS_ENC_UTF8_12_1, "utf8", UNICODE_AGE(12, 1, 0)}, }; -static int f2fs_sb_read_encoding(const struct f2fs_super_block *sb, - const struct f2fs_sb_encodings **encoding, - __u16 *flags) +static const struct f2fs_sb_encodings * +f2fs_sb_read_encoding(const struct f2fs_super_block *sb) { __u16 magic = le16_to_cpu(sb->s_encoding); int i; for (i = 0; i < ARRAY_SIZE(f2fs_sb_encoding_map); i++) if (magic == f2fs_sb_encoding_map[i].magic) - break; - - if (i >= ARRAY_SIZE(f2fs_sb_encoding_map)) - return -EINVAL; + return &f2fs_sb_encoding_map[i]; - *encoding = &f2fs_sb_encoding_map[i]; - *flags = le16_to_cpu(sb->s_encoding_flags); + return NULL; +} +struct kmem_cache *f2fs_cf_name_slab; +static int __init f2fs_create_casefold_cache(void) +{ + f2fs_cf_name_slab = f2fs_kmem_cache_create("f2fs_casefolded_name", + F2FS_NAME_LEN); + if (!f2fs_cf_name_slab) + return -ENOMEM; return 0; } + +static void f2fs_destroy_casefold_cache(void) +{ + kmem_cache_destroy(f2fs_cf_name_slab); +} +#else +static int __init f2fs_create_casefold_cache(void) { return 0; } +static void f2fs_destroy_casefold_cache(void) { } #endif static inline void limit_reserve_root(struct f2fs_sb_info *sbi) { - block_t limit = min((sbi->user_block_count << 1) / 1000, + block_t limit = min((sbi->user_block_count >> 3), sbi->user_block_count - sbi->reserved_blocks); - /* limit is 0.2% */ + /* limit is 12.5% */ if (test_opt(sbi, RESERVE_ROOT) && F2FS_OPTION(sbi).root_reserved_blocks > limit) { F2FS_OPTION(sbi).root_reserved_blocks = limit; @@ -283,6 +323,62 @@ static inline void limit_reserve_root(struct f2fs_sb_info *sbi) F2FS_OPTION(sbi).s_resgid)); } +static inline int adjust_reserved_segment(struct f2fs_sb_info *sbi) +{ + unsigned int sec_blks = sbi->blocks_per_seg * sbi->segs_per_sec; + unsigned int avg_vblocks; + unsigned int wanted_reserved_segments; + block_t avail_user_block_count; + + if (!F2FS_IO_ALIGNED(sbi)) + return 0; + + /* average valid block count in section in worst case */ + avg_vblocks = sec_blks / F2FS_IO_SIZE(sbi); + + /* + * we need enough free space when migrating one section in worst case + */ + wanted_reserved_segments = (F2FS_IO_SIZE(sbi) / avg_vblocks) * + reserved_segments(sbi); + wanted_reserved_segments -= reserved_segments(sbi); + + avail_user_block_count = sbi->user_block_count - + sbi->current_reserved_blocks - + F2FS_OPTION(sbi).root_reserved_blocks; + + if (wanted_reserved_segments * sbi->blocks_per_seg > + avail_user_block_count) { + f2fs_err(sbi, "IO align feature can't grab additional reserved segment: %u, available segments: %u", + wanted_reserved_segments, + avail_user_block_count >> sbi->log_blocks_per_seg); + return -ENOSPC; + } + + SM_I(sbi)->additional_reserved_segments = wanted_reserved_segments; + + f2fs_info(sbi, "IO align feature needs additional reserved segment: %u", + wanted_reserved_segments); + + return 0; +} + +static inline void adjust_unusable_cap_perc(struct f2fs_sb_info *sbi) +{ + if (!F2FS_OPTION(sbi).unusable_cap_perc) + return; + + if (F2FS_OPTION(sbi).unusable_cap_perc == 100) + F2FS_OPTION(sbi).unusable_cap = sbi->user_block_count; + else + F2FS_OPTION(sbi).unusable_cap = (sbi->user_block_count / 100) * + F2FS_OPTION(sbi).unusable_cap_perc; + + f2fs_info(sbi, "Adjust unusable cap for checkpoint=disable = %u / %u%%", + F2FS_OPTION(sbi).unusable_cap, + F2FS_OPTION(sbi).unusable_cap_perc); +} + static void init_once(void *foo) { struct f2fs_inode_info *fi = (struct f2fs_inode_info *) foo; @@ -330,7 +426,7 @@ static int f2fs_set_qf_name(struct super_block *sb, int qtype, set_opt(sbi, QUOTA); return 0; errout: - kvfree(qname); + kfree(qname); return ret; } @@ -342,7 +438,7 @@ static int f2fs_clear_qf_name(struct super_block *sb, int qtype) f2fs_err(sbi, "Cannot change journaled quota options when quota turned on"); return -EINVAL; } - kvfree(F2FS_OPTION(sbi).s_qf_names[qtype]); + kfree(F2FS_OPTION(sbi).s_qf_names[qtype]); F2FS_OPTION(sbi).s_qf_names[qtype] = NULL; return 0; } @@ -393,24 +489,190 @@ static int f2fs_check_quota_options(struct f2fs_sb_info *sbi) } #endif -static int parse_options(struct super_block *sb, char *options) +static int f2fs_set_test_dummy_encryption(struct super_block *sb, + const char *opt, + const substring_t *arg, + bool is_remount) +{ + struct f2fs_sb_info *sbi = F2FS_SB(sb); + struct fs_parameter param = { + .type = fs_value_is_string, + .string = arg->from ? arg->from : "", + }; + struct fscrypt_dummy_policy *policy = + &F2FS_OPTION(sbi).dummy_enc_policy; + int err; + + if (!IS_ENABLED(CONFIG_FS_ENCRYPTION)) { + f2fs_warn(sbi, "test_dummy_encryption option not supported"); + return -EINVAL; + } + + if (!f2fs_sb_has_encrypt(sbi)) { + f2fs_err(sbi, "Encrypt feature is off"); + return -EINVAL; + } + + /* + * This mount option is just for testing, and it's not worthwhile to + * implement the extra complexity (e.g. RCU protection) that would be + * needed to allow it to be set or changed during remount. We do allow + * it to be specified during remount, but only if there is no change. + */ + if (is_remount && !fscrypt_is_dummy_policy_set(policy)) { + f2fs_warn(sbi, "Can't set test_dummy_encryption on remount"); + return -EINVAL; + } + + err = fscrypt_parse_test_dummy_encryption(¶m, policy); + if (err) { + if (err == -EEXIST) + f2fs_warn(sbi, + "Can't change test_dummy_encryption on remount"); + else if (err == -EINVAL) + f2fs_warn(sbi, "Value of option \"%s\" is unrecognized", + opt); + else + f2fs_warn(sbi, "Error processing option \"%s\" [%d]", + opt, err); + return -EINVAL; + } + err = fscrypt_add_test_dummy_key(sb, policy); + if (err) { + f2fs_warn(sbi, "Error adding test dummy encryption key [%d]", + err); + return err; + } + f2fs_warn(sbi, "Test dummy encryption mode enabled"); + return 0; +} + +#ifdef CONFIG_F2FS_FS_COMPRESSION +/* + * 1. The same extension name cannot not appear in both compress and non-compress extension + * at the same time. + * 2. If the compress extension specifies all files, the types specified by the non-compress + * extension will be treated as special cases and will not be compressed. + * 3. Don't allow the non-compress extension specifies all files. + */ +static int f2fs_test_compress_extension(struct f2fs_sb_info *sbi) +{ + unsigned char (*ext)[F2FS_EXTENSION_LEN]; + unsigned char (*noext)[F2FS_EXTENSION_LEN]; + int ext_cnt, noext_cnt, index = 0, no_index = 0; + + ext = F2FS_OPTION(sbi).extensions; + ext_cnt = F2FS_OPTION(sbi).compress_ext_cnt; + noext = F2FS_OPTION(sbi).noextensions; + noext_cnt = F2FS_OPTION(sbi).nocompress_ext_cnt; + + if (!noext_cnt) + return 0; + + for (no_index = 0; no_index < noext_cnt; no_index++) { + if (!strcasecmp("*", noext[no_index])) { + f2fs_info(sbi, "Don't allow the nocompress extension specifies all files"); + return -EINVAL; + } + for (index = 0; index < ext_cnt; index++) { + if (!strcasecmp(ext[index], noext[no_index])) { + f2fs_info(sbi, "Don't allow the same extension %s appear in both compress and nocompress extension", + ext[index]); + return -EINVAL; + } + } + } + return 0; +} + +#ifdef CONFIG_F2FS_FS_LZ4 +static int f2fs_set_lz4hc_level(struct f2fs_sb_info *sbi, const char *str) +{ +#ifdef CONFIG_F2FS_FS_LZ4HC + unsigned int level; +#endif + + if (strlen(str) == 3) { + F2FS_OPTION(sbi).compress_level = 0; + return 0; + } + +#ifdef CONFIG_F2FS_FS_LZ4HC + str += 3; + + if (str[0] != ':') { + f2fs_info(sbi, "wrong format, e.g. <alg_name>:<compr_level>"); + return -EINVAL; + } + if (kstrtouint(str + 1, 10, &level)) + return -EINVAL; + + if (level < LZ4HC_MIN_CLEVEL || level > LZ4HC_MAX_CLEVEL) { + f2fs_info(sbi, "invalid lz4hc compress level: %d", level); + return -EINVAL; + } + + F2FS_OPTION(sbi).compress_level = level; + return 0; +#else + f2fs_info(sbi, "kernel doesn't support lz4hc compression"); + return -EINVAL; +#endif +} +#endif + +#ifdef CONFIG_F2FS_FS_ZSTD +static int f2fs_set_zstd_level(struct f2fs_sb_info *sbi, const char *str) +{ + unsigned int level; + int len = 4; + + if (strlen(str) == len) { + F2FS_OPTION(sbi).compress_level = 0; + return 0; + } + + str += len; + + if (str[0] != ':') { + f2fs_info(sbi, "wrong format, e.g. <alg_name>:<compr_level>"); + return -EINVAL; + } + if (kstrtouint(str + 1, 10, &level)) + return -EINVAL; + + if (!level || level > zstd_max_clevel()) { + f2fs_info(sbi, "invalid zstd compress level: %d", level); + return -EINVAL; + } + + F2FS_OPTION(sbi).compress_level = level; + return 0; +} +#endif +#endif + +static int parse_options(struct super_block *sb, char *options, bool is_remount) { struct f2fs_sb_info *sbi = F2FS_SB(sb); substring_t args[MAX_OPT_ARGS]; +#ifdef CONFIG_F2FS_FS_COMPRESSION unsigned char (*ext)[F2FS_EXTENSION_LEN]; + unsigned char (*noext)[F2FS_EXTENSION_LEN]; + int ext_cnt, noext_cnt; +#endif char *p, *name; - int arg = 0, ext_cnt; + int arg = 0; kuid_t uid; kgid_t gid; -#ifdef CONFIG_QUOTA int ret; -#endif if (!options) - return 0; + goto default_check; while ((p = strsep(&options, ",")) != NULL) { int token; + if (!*p) continue; /* @@ -426,35 +688,36 @@ static int parse_options(struct super_block *sb, char *options) if (!name) return -ENOMEM; - if (strlen(name) == 2 && !strncmp(name, "on", 2)) { - set_opt(sbi, BG_GC); - clear_opt(sbi, FORCE_FG_GC); - } else if (strlen(name) == 3 && !strncmp(name, "off", 3)) { - clear_opt(sbi, BG_GC); - clear_opt(sbi, FORCE_FG_GC); - } else if (strlen(name) == 4 && !strncmp(name, "sync", 4)) { - set_opt(sbi, BG_GC); - set_opt(sbi, FORCE_FG_GC); + if (!strcmp(name, "on")) { + F2FS_OPTION(sbi).bggc_mode = BGGC_MODE_ON; + } else if (!strcmp(name, "off")) { + F2FS_OPTION(sbi).bggc_mode = BGGC_MODE_OFF; + } else if (!strcmp(name, "sync")) { + F2FS_OPTION(sbi).bggc_mode = BGGC_MODE_SYNC; } else { - kvfree(name); + kfree(name); return -EINVAL; } - kvfree(name); + kfree(name); break; case Opt_disable_roll_forward: set_opt(sbi, DISABLE_ROLL_FORWARD); break; case Opt_norecovery: /* this option mounts f2fs with ro */ - set_opt(sbi, DISABLE_ROLL_FORWARD); + set_opt(sbi, NORECOVERY); if (!f2fs_readonly(sb)) return -EINVAL; break; case Opt_discard: + if (!f2fs_hw_support_discard(sbi)) { + f2fs_warn(sbi, "device does not support discard"); + break; + } set_opt(sbi, DISCARD); break; case Opt_nodiscard: - if (f2fs_sb_has_blkzoned(sbi)) { + if (f2fs_hw_should_discard(sbi)) { f2fs_warn(sbi, "discard is required for zoned block devices"); return -EINVAL; } @@ -517,7 +780,8 @@ static int parse_options(struct super_block *sb, char *options) case Opt_active_logs: if (args->from && match_int(args, &arg)) return -EINVAL; - if (arg != 2 && arg != 4 && arg != NR_CURSEG_TYPE) + if (arg != 2 && arg != 4 && + arg != NR_CURSEG_PERSIST_TYPE) return -EINVAL; F2FS_OPTION(sbi).active_logs = arg; break; @@ -593,29 +857,31 @@ static int parse_options(struct super_block *sb, char *options) if (!name) return -ENOMEM; - if (strlen(name) == 8 && - !strncmp(name, "adaptive", 8)) { + if (!strcmp(name, "adaptive")) { if (f2fs_sb_has_blkzoned(sbi)) { f2fs_warn(sbi, "adaptive mode is not allowed with zoned block device feature"); - kvfree(name); + kfree(name); return -EINVAL; } - set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE); - } else if (strlen(name) == 3 && - !strncmp(name, "lfs", 3)) { - set_opt_mode(sbi, F2FS_MOUNT_LFS); + F2FS_OPTION(sbi).fs_mode = FS_MODE_ADAPTIVE; + } else if (!strcmp(name, "lfs")) { + F2FS_OPTION(sbi).fs_mode = FS_MODE_LFS; + } else if (!strcmp(name, "fragment:segment")) { + F2FS_OPTION(sbi).fs_mode = FS_MODE_FRAGMENT_SEG; + } else if (!strcmp(name, "fragment:block")) { + F2FS_OPTION(sbi).fs_mode = FS_MODE_FRAGMENT_BLK; } else { - kvfree(name); + kfree(name); return -EINVAL; } - kvfree(name); + kfree(name); break; case Opt_io_size_bits: if (args->from && match_int(args, &arg)) return -EINVAL; - if (arg <= 0 || arg > __ilog2_u32(BIO_MAX_PAGES)) { + if (arg <= 0 || arg > __ilog2_u32(BIO_MAX_VECS)) { f2fs_warn(sbi, "Not support %d, larger than %d", - 1 << arg, BIO_MAX_PAGES); + 1 << arg, BIO_MAX_VECS); return -EINVAL; } F2FS_OPTION(sbi).write_io_size_bits = arg; @@ -723,73 +989,49 @@ static int parse_options(struct super_block *sb, char *options) f2fs_info(sbi, "quota operations not supported"); break; #endif - case Opt_whint: - name = match_strdup(&args[0]); - if (!name) - return -ENOMEM; - if (strlen(name) == 10 && - !strncmp(name, "user-based", 10)) { - F2FS_OPTION(sbi).whint_mode = WHINT_MODE_USER; - } else if (strlen(name) == 3 && - !strncmp(name, "off", 3)) { - F2FS_OPTION(sbi).whint_mode = WHINT_MODE_OFF; - } else if (strlen(name) == 8 && - !strncmp(name, "fs-based", 8)) { - F2FS_OPTION(sbi).whint_mode = WHINT_MODE_FS; - } else { - kvfree(name); - return -EINVAL; - } - kvfree(name); - break; case Opt_alloc: name = match_strdup(&args[0]); if (!name) return -ENOMEM; - if (strlen(name) == 7 && - !strncmp(name, "default", 7)) { + if (!strcmp(name, "default")) { F2FS_OPTION(sbi).alloc_mode = ALLOC_MODE_DEFAULT; - } else if (strlen(name) == 5 && - !strncmp(name, "reuse", 5)) { + } else if (!strcmp(name, "reuse")) { F2FS_OPTION(sbi).alloc_mode = ALLOC_MODE_REUSE; } else { - kvfree(name); + kfree(name); return -EINVAL; } - kvfree(name); + kfree(name); break; case Opt_fsync: name = match_strdup(&args[0]); if (!name) return -ENOMEM; - if (strlen(name) == 5 && - !strncmp(name, "posix", 5)) { + if (!strcmp(name, "posix")) { F2FS_OPTION(sbi).fsync_mode = FSYNC_MODE_POSIX; - } else if (strlen(name) == 6 && - !strncmp(name, "strict", 6)) { + } else if (!strcmp(name, "strict")) { F2FS_OPTION(sbi).fsync_mode = FSYNC_MODE_STRICT; - } else if (strlen(name) == 9 && - !strncmp(name, "nobarrier", 9)) { + } else if (!strcmp(name, "nobarrier")) { F2FS_OPTION(sbi).fsync_mode = FSYNC_MODE_NOBARRIER; } else { - kvfree(name); + kfree(name); return -EINVAL; } - kvfree(name); + kfree(name); break; case Opt_test_dummy_encryption: -#ifdef CONFIG_FS_ENCRYPTION - if (!f2fs_sb_has_encrypt(sbi)) { - f2fs_err(sbi, "Encrypt feature is off"); - return -EINVAL; - } - - F2FS_OPTION(sbi).test_dummy_encryption = true; - f2fs_info(sbi, "Test dummy encryption mode enabled"); + ret = f2fs_set_test_dummy_encryption(sb, p, &args[0], + is_remount); + if (ret) + return ret; + break; + case Opt_inlinecrypt: +#ifdef CONFIG_FS_ENCRYPTION_INLINE_CRYPT + sb->s_flags |= SB_INLINECRYPT; #else - f2fs_info(sbi, "Test dummy encryption mount option ignored"); + f2fs_info(sbi, "inline encryption not supported"); #endif break; case Opt_checkpoint_disable_cap_perc: @@ -797,12 +1039,7 @@ static int parse_options(struct super_block *sb, char *options) return -EINVAL; if (arg < 0 || arg > 100) return -EINVAL; - if (arg == 100) - F2FS_OPTION(sbi).unusable_cap = - sbi->user_block_count; - else - F2FS_OPTION(sbi).unusable_cap = - (sbi->user_block_count / 100) * arg; + F2FS_OPTION(sbi).unusable_cap_perc = arg; set_opt(sbi, DISABLE_CHECKPOINT); break; case Opt_checkpoint_disable_cap: @@ -817,21 +1054,61 @@ static int parse_options(struct super_block *sb, char *options) case Opt_checkpoint_enable: clear_opt(sbi, DISABLE_CHECKPOINT); break; + case Opt_checkpoint_merge: + set_opt(sbi, MERGE_CHECKPOINT); + break; + case Opt_nocheckpoint_merge: + clear_opt(sbi, MERGE_CHECKPOINT); + break; +#ifdef CONFIG_F2FS_FS_COMPRESSION case Opt_compress_algorithm: if (!f2fs_sb_has_compression(sbi)) { - f2fs_err(sbi, "Compression feature if off"); - return -EINVAL; + f2fs_info(sbi, "Image doesn't support compression"); + break; } name = match_strdup(&args[0]); if (!name) return -ENOMEM; - if (strlen(name) == 3 && !strcmp(name, "lzo")) { + if (!strcmp(name, "lzo")) { +#ifdef CONFIG_F2FS_FS_LZO + F2FS_OPTION(sbi).compress_level = 0; F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZO; - } else if (strlen(name) == 3 && - !strcmp(name, "lz4")) { +#else + f2fs_info(sbi, "kernel doesn't support lzo compression"); +#endif + } else if (!strncmp(name, "lz4", 3)) { +#ifdef CONFIG_F2FS_FS_LZ4 + ret = f2fs_set_lz4hc_level(sbi, name); + if (ret) { + kfree(name); + return -EINVAL; + } F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZ4; +#else + f2fs_info(sbi, "kernel doesn't support lz4 compression"); +#endif + } else if (!strncmp(name, "zstd", 4)) { +#ifdef CONFIG_F2FS_FS_ZSTD + ret = f2fs_set_zstd_level(sbi, name); + if (ret) { + kfree(name); + return -EINVAL; + } + F2FS_OPTION(sbi).compress_algorithm = + COMPRESS_ZSTD; +#else + f2fs_info(sbi, "kernel doesn't support zstd compression"); +#endif + } else if (!strcmp(name, "lzo-rle")) { +#ifdef CONFIG_F2FS_FS_LZORLE + F2FS_OPTION(sbi).compress_level = 0; + F2FS_OPTION(sbi).compress_algorithm = + COMPRESS_LZORLE; +#else + f2fs_info(sbi, "kernel doesn't support lzorle compression"); +#endif } else { kfree(name); return -EINVAL; @@ -840,8 +1117,8 @@ static int parse_options(struct super_block *sb, char *options) break; case Opt_compress_log_size: if (!f2fs_sb_has_compression(sbi)) { - f2fs_err(sbi, "Compression feature is off"); - return -EINVAL; + f2fs_info(sbi, "Image doesn't support compression"); + break; } if (args->from && match_int(args, &arg)) return -EINVAL; @@ -855,8 +1132,8 @@ static int parse_options(struct super_block *sb, char *options) break; case Opt_compress_extension: if (!f2fs_sb_has_compression(sbi)) { - f2fs_err(sbi, "Compression feature is off"); - return -EINVAL; + f2fs_info(sbi, "Image doesn't support compression"); + break; } name = match_strdup(&args[0]); if (!name) @@ -877,12 +1154,112 @@ static int parse_options(struct super_block *sb, char *options) F2FS_OPTION(sbi).compress_ext_cnt++; kfree(name); break; + case Opt_nocompress_extension: + if (!f2fs_sb_has_compression(sbi)) { + f2fs_info(sbi, "Image doesn't support compression"); + break; + } + name = match_strdup(&args[0]); + if (!name) + return -ENOMEM; + + noext = F2FS_OPTION(sbi).noextensions; + noext_cnt = F2FS_OPTION(sbi).nocompress_ext_cnt; + + if (strlen(name) >= F2FS_EXTENSION_LEN || + noext_cnt >= COMPRESS_EXT_NUM) { + f2fs_err(sbi, + "invalid extension length/number"); + kfree(name); + return -EINVAL; + } + + strcpy(noext[noext_cnt], name); + F2FS_OPTION(sbi).nocompress_ext_cnt++; + kfree(name); + break; + case Opt_compress_chksum: + F2FS_OPTION(sbi).compress_chksum = true; + break; + case Opt_compress_mode: + name = match_strdup(&args[0]); + if (!name) + return -ENOMEM; + if (!strcmp(name, "fs")) { + F2FS_OPTION(sbi).compress_mode = COMPR_MODE_FS; + } else if (!strcmp(name, "user")) { + F2FS_OPTION(sbi).compress_mode = COMPR_MODE_USER; + } else { + kfree(name); + return -EINVAL; + } + kfree(name); + break; + case Opt_compress_cache: + set_opt(sbi, COMPRESS_CACHE); + break; +#else + case Opt_compress_algorithm: + case Opt_compress_log_size: + case Opt_compress_extension: + case Opt_nocompress_extension: + case Opt_compress_chksum: + case Opt_compress_mode: + case Opt_compress_cache: + f2fs_info(sbi, "compression options not supported"); + break; +#endif + case Opt_atgc: + set_opt(sbi, ATGC); + break; + case Opt_gc_merge: + set_opt(sbi, GC_MERGE); + break; + case Opt_nogc_merge: + clear_opt(sbi, GC_MERGE); + break; + case Opt_discard_unit: + name = match_strdup(&args[0]); + if (!name) + return -ENOMEM; + if (!strcmp(name, "block")) { + F2FS_OPTION(sbi).discard_unit = + DISCARD_UNIT_BLOCK; + } else if (!strcmp(name, "segment")) { + F2FS_OPTION(sbi).discard_unit = + DISCARD_UNIT_SEGMENT; + } else if (!strcmp(name, "section")) { + F2FS_OPTION(sbi).discard_unit = + DISCARD_UNIT_SECTION; + } else { + kfree(name); + return -EINVAL; + } + kfree(name); + break; + case Opt_memory_mode: + name = match_strdup(&args[0]); + if (!name) + return -ENOMEM; + if (!strcmp(name, "normal")) { + F2FS_OPTION(sbi).memory_mode = + MEMORY_MODE_NORMAL; + } else if (!strcmp(name, "low")) { + F2FS_OPTION(sbi).memory_mode = + MEMORY_MODE_LOW; + } else { + kfree(name); + return -EINVAL; + } + kfree(name); + break; default: f2fs_err(sbi, "Unrecognized mount option \"%s\" or missing value", p); return -EINVAL; } } +default_check: #ifdef CONFIG_QUOTA if (f2fs_check_quota_options(sbi)) return -EINVAL; @@ -896,15 +1273,41 @@ static int parse_options(struct super_block *sb, char *options) return -EINVAL; } #endif -#ifndef CONFIG_UNICODE +#if !IS_ENABLED(CONFIG_UNICODE) if (f2fs_sb_has_casefold(sbi)) { f2fs_err(sbi, "Filesystem with casefold feature cannot be mounted without CONFIG_UNICODE"); return -EINVAL; } #endif + /* + * The BLKZONED feature indicates that the drive was formatted with + * zone alignment optimization. This is optional for host-aware + * devices, but mandatory for host-managed zoned block devices. + */ +#ifndef CONFIG_BLK_DEV_ZONED + if (f2fs_sb_has_blkzoned(sbi)) { + f2fs_err(sbi, "Zoned block device support is not enabled"); + return -EINVAL; + } +#endif + if (f2fs_sb_has_blkzoned(sbi)) { + if (F2FS_OPTION(sbi).discard_unit != + DISCARD_UNIT_SECTION) { + f2fs_info(sbi, "Zoned block device doesn't need small discard, set discard_unit=section by default"); + F2FS_OPTION(sbi).discard_unit = + DISCARD_UNIT_SECTION; + } + } - if (F2FS_IO_SIZE_BITS(sbi) && !test_opt(sbi, LFS)) { +#ifdef CONFIG_F2FS_FS_COMPRESSION + if (f2fs_test_compress_extension(sbi)) { + f2fs_err(sbi, "invalid compress or nocompress extension"); + return -EINVAL; + } +#endif + + if (F2FS_IO_SIZE_BITS(sbi) && !f2fs_lfs_mode(sbi)) { f2fs_err(sbi, "Should set mode=lfs with %uKB-sized IO", F2FS_IO_SIZE_KB(sbi)); return -EINVAL; @@ -934,16 +1337,20 @@ static int parse_options(struct super_block *sb, char *options) } } - if (test_opt(sbi, DISABLE_CHECKPOINT) && test_opt(sbi, LFS)) { - f2fs_err(sbi, "LFS not compatible with checkpoint=disable\n"); + if (test_opt(sbi, DISABLE_CHECKPOINT) && f2fs_lfs_mode(sbi)) { + f2fs_err(sbi, "LFS not compatible with checkpoint=disable"); return -EINVAL; } - /* Not pass down write hints if the number of active logs is lesser - * than NR_CURSEG_TYPE. - */ - if (F2FS_OPTION(sbi).active_logs != NR_CURSEG_TYPE) - F2FS_OPTION(sbi).whint_mode = WHINT_MODE_OFF; + if (test_opt(sbi, ATGC) && f2fs_lfs_mode(sbi)) { + f2fs_err(sbi, "LFS not compatible with ATGC"); + return -EINVAL; + } + + if (f2fs_sb_has_readonly(sbi) && !f2fs_readonly(sbi->sb)) { + f2fs_err(sbi, "Allow to mount readonly mode only"); + return -EROFS; + } return 0; } @@ -951,7 +1358,12 @@ static struct inode *f2fs_alloc_inode(struct super_block *sb) { struct f2fs_inode_info *fi; - fi = kmem_cache_alloc(f2fs_inode_cachep, GFP_F2FS_ZERO); + if (time_to_inject(F2FS_SB(sb), FAULT_SLAB_ALLOC)) { + f2fs_show_injection_info(F2FS_SB(sb), FAULT_SLAB_ALLOC); + return NULL; + } + + fi = alloc_inode_sb(sb, f2fs_inode_cachep, GFP_F2FS_ZERO); if (!fi) return NULL; @@ -959,16 +1371,14 @@ static struct inode *f2fs_alloc_inode(struct super_block *sb) /* Initialize f2fs-specific inode info */ atomic_set(&fi->dirty_pages, 0); - init_rwsem(&fi->i_sem); + atomic_set(&fi->i_compr_blocks, 0); + init_f2fs_rwsem(&fi->i_sem); + spin_lock_init(&fi->i_size_lock); INIT_LIST_HEAD(&fi->dirty_list); INIT_LIST_HEAD(&fi->gdirty_list); - INIT_LIST_HEAD(&fi->inmem_ilist); - INIT_LIST_HEAD(&fi->inmem_pages); - mutex_init(&fi->inmem_lock); - init_rwsem(&fi->i_gc_rwsem[READ]); - init_rwsem(&fi->i_gc_rwsem[WRITE]); - init_rwsem(&fi->i_mmap_sem); - init_rwsem(&fi->i_xattr_sem); + init_f2fs_rwsem(&fi->i_gc_rwsem[READ]); + init_f2fs_rwsem(&fi->i_gc_rwsem[WRITE]); + init_f2fs_rwsem(&fi->i_xattr_sem); /* Will be used by directory only */ fi->i_dir_level = F2FS_SB(sb)->dir_level; @@ -1006,9 +1416,7 @@ static int f2fs_drop_inode(struct inode *inode) atomic_inc(&inode->i_count); spin_unlock(&inode->i_lock); - /* some remained atomic pages should discarded */ - if (f2fs_is_atomic_file(inode)) - f2fs_drop_inmem_pages(inode); + f2fs_abort_atomic_write(inode, true); /* should remain fi->extent_tree for writepage */ f2fs_destroy_extent_node(inode); @@ -1091,9 +1499,6 @@ static void f2fs_dirty_inode(struct inode *inode, int flags) inode->i_ino == F2FS_META_INO(sbi)) return; - if (flags == I_DIRTY_TIME) - return; - if (is_inode_flag_set(inode, FI_AUTO_RECOVER)) clear_inode_flag(inode, FI_AUTO_RECOVER); @@ -1108,8 +1513,9 @@ static void f2fs_free_inode(struct inode *inode) static void destroy_percpu_info(struct f2fs_sb_info *sbi) { - percpu_counter_destroy(&sbi->alloc_valid_block_count); percpu_counter_destroy(&sbi->total_valid_inode_count); + percpu_counter_destroy(&sbi->rf_node_block_count); + percpu_counter_destroy(&sbi->alloc_valid_block_count); } static void destroy_device_list(struct f2fs_sb_info *sbi) @@ -1131,12 +1537,21 @@ static void f2fs_put_super(struct super_block *sb) int i; bool dropped; + /* unregister procfs/sysfs entries in advance to avoid race case */ + f2fs_unregister_sysfs(sbi); + f2fs_quota_off_umount(sb); /* prevent remaining shrinker jobs */ mutex_lock(&sbi->umount_mutex); /* + * flush all issued checkpoints and stop checkpoint issue thread. + * after then, all checkpoints should be done by each process context. + */ + f2fs_stop_ckpt_thread(sbi); + + /* * We don't need to do checkpoint when superblock is clean. * But, the previous checkpoint was not done by umount, it needs to do * clean checkpoint again. @@ -1172,10 +1587,12 @@ static void f2fs_put_super(struct super_block *sb) /* our cp_error case, we can wait for any writeback page */ f2fs_flush_merged_writes(sbi); - f2fs_wait_on_all_pages_writeback(sbi); + f2fs_wait_on_all_pages(sbi, F2FS_WB_CP_DATA); f2fs_bug_on(sbi, sbi->fsync_node_num); + f2fs_destroy_compress_inode(sbi); + iput(sbi->node_inode); sbi->node_inode = NULL; @@ -1196,26 +1613,28 @@ static void f2fs_put_super(struct super_block *sb) kvfree(sbi->ckpt); - f2fs_unregister_sysfs(sbi); - sb->s_fs_info = NULL; if (sbi->s_chksum_driver) crypto_free_shash(sbi->s_chksum_driver); - kvfree(sbi->raw_super); + kfree(sbi->raw_super); destroy_device_list(sbi); + f2fs_destroy_page_array_cache(sbi); + f2fs_destroy_xattr_caches(sbi); mempool_destroy(sbi->write_io_dummy); #ifdef CONFIG_QUOTA for (i = 0; i < MAXQUOTAS; i++) - kvfree(F2FS_OPTION(sbi).s_qf_names[i]); + kfree(F2FS_OPTION(sbi).s_qf_names[i]); #endif + fscrypt_free_dummy_policy(&F2FS_OPTION(sbi).dummy_enc_policy); destroy_percpu_info(sbi); + f2fs_destroy_iostat(sbi); for (i = 0; i < NR_PAGE_TYPE; i++) kvfree(sbi->write_io[i]); -#ifdef CONFIG_UNICODE - utf8_unload(sbi->s_encoding); +#if IS_ENABLED(CONFIG_UNICODE) + utf8_unload(sb->s_encoding); #endif - kvfree(sbi); + kfree(sbi); } int f2fs_sync_fs(struct super_block *sb, int sync) @@ -1233,16 +1652,8 @@ int f2fs_sync_fs(struct super_block *sb, int sync) if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING))) return -EAGAIN; - if (sync) { - struct cp_control cpc; - - cpc.reason = __get_cp_reason(sbi); - - down_write(&sbi->gc_lock); - err = f2fs_write_checkpoint(sbi, &cpc); - up_write(&sbi->gc_lock); - } - f2fs_trace_ios(NULL, 1); + if (sync) + err = f2fs_issue_checkpoint(sbi); return err; } @@ -1259,11 +1670,18 @@ static int f2fs_freeze(struct super_block *sb) /* must be clean, since sync_filesystem() was already called */ if (is_sbi_flag_set(F2FS_SB(sb), SBI_IS_DIRTY)) return -EINVAL; + + /* Let's flush checkpoints and stop the thread. */ + f2fs_flush_ckpt_thread(F2FS_SB(sb)); + + /* to avoid deadlock on f2fs_evict_inode->SB_FREEZE_FS */ + set_sbi_flag(F2FS_SB(sb), SBI_IS_FREEZING); return 0; } static int f2fs_unfreeze(struct super_block *sb) { + clear_sbi_flag(F2FS_SB(sb), SBI_IS_FREEZING); return 0; } @@ -1288,7 +1706,8 @@ static int f2fs_statfs_project(struct super_block *sb, limit >>= sb->s_blocksize_bits; if (limit && buf->f_blocks > limit) { - curblock = dquot->dq_dqb.dqb_curspace >> sb->s_blocksize_bits; + curblock = (dquot->dq_dqb.dqb_curspace + + dquot->dq_dqb.dqb_rsvspace) >> sb->s_blocksize_bits; buf->f_blocks = limit; buf->f_bfree = buf->f_bavail = (buf->f_blocks > curblock) ? @@ -1318,18 +1737,23 @@ static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf) u64 id = huge_encode_dev(sb->s_bdev->bd_dev); block_t total_count, user_block_count, start_count; u64 avail_node_count; + unsigned int total_valid_node_count; total_count = le64_to_cpu(sbi->raw_super->block_count); - user_block_count = sbi->user_block_count; start_count = le32_to_cpu(sbi->raw_super->segment0_blkaddr); buf->f_type = F2FS_SUPER_MAGIC; buf->f_bsize = sbi->blocksize; buf->f_blocks = total_count - start_count; + + spin_lock(&sbi->stat_lock); + + user_block_count = sbi->user_block_count; + total_valid_node_count = valid_node_count(sbi); + avail_node_count = sbi->total_node_count - F2FS_RESERVED_NODE_NUM; buf->f_bfree = user_block_count - valid_user_blocks(sbi) - sbi->current_reserved_blocks; - spin_lock(&sbi->stat_lock); if (unlikely(buf->f_bfree <= sbi->unusable_block_count)) buf->f_bfree = 0; else @@ -1342,20 +1766,17 @@ static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf) else buf->f_bavail = 0; - avail_node_count = sbi->total_node_count - F2FS_RESERVED_NODE_NUM; - if (avail_node_count > user_block_count) { buf->f_files = user_block_count; buf->f_ffree = buf->f_bavail; } else { buf->f_files = avail_node_count; - buf->f_ffree = min(avail_node_count - valid_node_count(sbi), + buf->f_ffree = min(avail_node_count - total_valid_node_count, buf->f_bavail); } buf->f_namelen = F2FS_NAME_LEN; - buf->f_fsid.val[0] = (u32)id; - buf->f_fsid.val[1] = (u32)(id >> 32); + buf->f_fsid = u64_to_fsid(id); #ifdef CONFIG_QUOTA if (is_inode_flag_set(dentry->d_inode, FI_PROJ_INHERIT) && @@ -1403,6 +1824,7 @@ static inline void f2fs_show_quota_options(struct seq_file *seq, #endif } +#ifdef CONFIG_F2FS_FS_COMPRESSION static inline void f2fs_show_compress_options(struct seq_file *seq, struct super_block *sb) { @@ -1420,9 +1842,18 @@ static inline void f2fs_show_compress_options(struct seq_file *seq, case COMPRESS_LZ4: algtype = "lz4"; break; + case COMPRESS_ZSTD: + algtype = "zstd"; + break; + case COMPRESS_LZORLE: + algtype = "lzo-rle"; + break; } seq_printf(seq, ",compress_algorithm=%s", algtype); + if (F2FS_OPTION(sbi).compress_level) + seq_printf(seq, ":%d", F2FS_OPTION(sbi).compress_level); + seq_printf(seq, ",compress_log_size=%u", F2FS_OPTION(sbi).compress_log_size); @@ -1430,22 +1861,43 @@ static inline void f2fs_show_compress_options(struct seq_file *seq, seq_printf(seq, ",compress_extension=%s", F2FS_OPTION(sbi).extensions[i]); } + + for (i = 0; i < F2FS_OPTION(sbi).nocompress_ext_cnt; i++) { + seq_printf(seq, ",nocompress_extension=%s", + F2FS_OPTION(sbi).noextensions[i]); + } + + if (F2FS_OPTION(sbi).compress_chksum) + seq_puts(seq, ",compress_chksum"); + + if (F2FS_OPTION(sbi).compress_mode == COMPR_MODE_FS) + seq_printf(seq, ",compress_mode=%s", "fs"); + else if (F2FS_OPTION(sbi).compress_mode == COMPR_MODE_USER) + seq_printf(seq, ",compress_mode=%s", "user"); + + if (test_opt(sbi, COMPRESS_CACHE)) + seq_puts(seq, ",compress_cache"); } +#endif static int f2fs_show_options(struct seq_file *seq, struct dentry *root) { struct f2fs_sb_info *sbi = F2FS_SB(root->d_sb); - if (!f2fs_readonly(sbi->sb) && test_opt(sbi, BG_GC)) { - if (test_opt(sbi, FORCE_FG_GC)) - seq_printf(seq, ",background_gc=%s", "sync"); - else - seq_printf(seq, ",background_gc=%s", "on"); - } else { + if (F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_SYNC) + seq_printf(seq, ",background_gc=%s", "sync"); + else if (F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_ON) + seq_printf(seq, ",background_gc=%s", "on"); + else if (F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_OFF) seq_printf(seq, ",background_gc=%s", "off"); - } + + if (test_opt(sbi, GC_MERGE)) + seq_puts(seq, ",gc_merge"); + if (test_opt(sbi, DISABLE_ROLL_FORWARD)) seq_puts(seq, ",disable_roll_forward"); + if (test_opt(sbi, NORECOVERY)) + seq_puts(seq, ",norecovery"); if (test_opt(sbi, DISCARD)) seq_puts(seq, ",discard"); else @@ -1497,10 +1949,14 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) seq_puts(seq, ",data_flush"); seq_puts(seq, ",mode="); - if (test_opt(sbi, ADAPTIVE)) + if (F2FS_OPTION(sbi).fs_mode == FS_MODE_ADAPTIVE) seq_puts(seq, "adaptive"); - else if (test_opt(sbi, LFS)) + else if (F2FS_OPTION(sbi).fs_mode == FS_MODE_LFS) seq_puts(seq, "lfs"); + else if (F2FS_OPTION(sbi).fs_mode == FS_MODE_FRAGMENT_SEG) + seq_puts(seq, "fragment:segment"); + else if (F2FS_OPTION(sbi).fs_mode == FS_MODE_FRAGMENT_BLK) + seq_puts(seq, "fragment:block"); seq_printf(seq, ",active_logs=%u", F2FS_OPTION(sbi).active_logs); if (test_opt(sbi, RESERVE_ROOT)) seq_printf(seq, ",reserve_root=%u,resuid=%u,resgid=%u", @@ -1531,14 +1987,11 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) seq_puts(seq, ",prjquota"); #endif f2fs_show_quota_options(seq, sbi->sb); - if (F2FS_OPTION(sbi).whint_mode == WHINT_MODE_USER) - seq_printf(seq, ",whint_mode=%s", "user-based"); - else if (F2FS_OPTION(sbi).whint_mode == WHINT_MODE_FS) - seq_printf(seq, ",whint_mode=%s", "fs-based"); -#ifdef CONFIG_FS_ENCRYPTION - if (F2FS_OPTION(sbi).test_dummy_encryption) - seq_puts(seq, ",test_dummy_encryption"); -#endif + + fscrypt_show_test_dummy_encryption(seq, ',', sbi->sb); + + if (sbi->sb->s_flags & SB_INLINECRYPT) + seq_puts(seq, ",inlinecrypt"); if (F2FS_OPTION(sbi).alloc_mode == ALLOC_MODE_DEFAULT) seq_printf(seq, ",alloc_mode=%s", "default"); @@ -1548,6 +2001,10 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) if (test_opt(sbi, DISABLE_CHECKPOINT)) seq_printf(seq, ",checkpoint=disable:%u", F2FS_OPTION(sbi).unusable_cap); + if (test_opt(sbi, MERGE_CHECKPOINT)) + seq_puts(seq, ",checkpoint_merge"); + else + seq_puts(seq, ",nocheckpoint_merge"); if (F2FS_OPTION(sbi).fsync_mode == FSYNC_MODE_POSIX) seq_printf(seq, ",fsync_mode=%s", "posix"); else if (F2FS_OPTION(sbi).fsync_mode == FSYNC_MODE_STRICT) @@ -1555,40 +2012,69 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) else if (F2FS_OPTION(sbi).fsync_mode == FSYNC_MODE_NOBARRIER) seq_printf(seq, ",fsync_mode=%s", "nobarrier"); +#ifdef CONFIG_F2FS_FS_COMPRESSION f2fs_show_compress_options(seq, sbi->sb); +#endif + + if (test_opt(sbi, ATGC)) + seq_puts(seq, ",atgc"); + + if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_BLOCK) + seq_printf(seq, ",discard_unit=%s", "block"); + else if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SEGMENT) + seq_printf(seq, ",discard_unit=%s", "segment"); + else if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SECTION) + seq_printf(seq, ",discard_unit=%s", "section"); + + if (F2FS_OPTION(sbi).memory_mode == MEMORY_MODE_NORMAL) + seq_printf(seq, ",memory=%s", "normal"); + else if (F2FS_OPTION(sbi).memory_mode == MEMORY_MODE_LOW) + seq_printf(seq, ",memory=%s", "low"); + return 0; } static void default_options(struct f2fs_sb_info *sbi) { /* init some FS parameters */ - F2FS_OPTION(sbi).active_logs = NR_CURSEG_TYPE; + if (f2fs_sb_has_readonly(sbi)) + F2FS_OPTION(sbi).active_logs = NR_CURSEG_RO_TYPE; + else + F2FS_OPTION(sbi).active_logs = NR_CURSEG_PERSIST_TYPE; + F2FS_OPTION(sbi).inline_xattr_size = DEFAULT_INLINE_XATTR_ADDRS; - F2FS_OPTION(sbi).whint_mode = WHINT_MODE_OFF; F2FS_OPTION(sbi).alloc_mode = ALLOC_MODE_DEFAULT; F2FS_OPTION(sbi).fsync_mode = FSYNC_MODE_POSIX; - F2FS_OPTION(sbi).test_dummy_encryption = false; F2FS_OPTION(sbi).s_resuid = make_kuid(&init_user_ns, F2FS_DEF_RESUID); F2FS_OPTION(sbi).s_resgid = make_kgid(&init_user_ns, F2FS_DEF_RESGID); - F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZO; + F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZ4; F2FS_OPTION(sbi).compress_log_size = MIN_COMPRESS_LOG_SIZE; F2FS_OPTION(sbi).compress_ext_cnt = 0; + F2FS_OPTION(sbi).compress_mode = COMPR_MODE_FS; + F2FS_OPTION(sbi).bggc_mode = BGGC_MODE_ON; + F2FS_OPTION(sbi).memory_mode = MEMORY_MODE_NORMAL; + + sbi->sb->s_flags &= ~SB_INLINECRYPT; - set_opt(sbi, BG_GC); set_opt(sbi, INLINE_XATTR); set_opt(sbi, INLINE_DATA); set_opt(sbi, INLINE_DENTRY); set_opt(sbi, EXTENT_CACHE); set_opt(sbi, NOHEAP); clear_opt(sbi, DISABLE_CHECKPOINT); + set_opt(sbi, MERGE_CHECKPOINT); F2FS_OPTION(sbi).unusable_cap = 0; sbi->sb->s_flags |= SB_LAZYTIME; set_opt(sbi, FLUSH_MERGE); - set_opt(sbi, DISCARD); - if (f2fs_sb_has_blkzoned(sbi)) - set_opt_mode(sbi, F2FS_MOUNT_LFS); - else - set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE); + if (f2fs_hw_support_discard(sbi) || f2fs_hw_should_discard(sbi)) + set_opt(sbi, DISCARD); + if (f2fs_sb_has_blkzoned(sbi)) { + F2FS_OPTION(sbi).fs_mode = FS_MODE_LFS; + F2FS_OPTION(sbi).discard_unit = DISCARD_UNIT_SECTION; + } else { + F2FS_OPTION(sbi).fs_mode = FS_MODE_ADAPTIVE; + F2FS_OPTION(sbi).discard_unit = DISCARD_UNIT_BLOCK; + } #ifdef CONFIG_F2FS_FS_XATTR set_opt(sbi, XATTR_USER); @@ -1608,6 +2094,7 @@ static int f2fs_disable_checkpoint(struct f2fs_sb_info *sbi) { unsigned int s_flags = sbi->sb->s_flags; struct cp_control cpc; + unsigned int gc_mode = sbi->gc_mode; int err = 0; int ret; block_t unusable; @@ -1618,11 +2105,25 @@ static int f2fs_disable_checkpoint(struct f2fs_sb_info *sbi) } sbi->sb->s_flags |= SB_ACTIVE; + /* check if we need more GC first */ + unusable = f2fs_get_unusable_blocks(sbi); + if (!f2fs_disable_cp_again(sbi, unusable)) + goto skip_gc; + f2fs_update_time(sbi, DISABLE_TIME); + sbi->gc_mode = GC_URGENT_HIGH; + while (!f2fs_time_over(sbi, DISABLE_TIME)) { - down_write(&sbi->gc_lock); - err = f2fs_gc(sbi, true, false, NULL_SEGNO); + struct f2fs_gc_control gc_control = { + .victim_segno = NULL_SEGNO, + .init_gc_type = FG_GC, + .should_migrate_blocks = false, + .err_gc_skipped = true, + .nr_free_secs = 1 }; + + f2fs_down_write(&sbi->gc_lock); + err = f2fs_gc(sbi, &gc_control); if (err == -ENODATA) { err = 0; break; @@ -1633,7 +2134,7 @@ static int f2fs_disable_checkpoint(struct f2fs_sb_info *sbi) ret = sync_filesystem(sbi->sb); if (ret || err) { - err = ret ? ret: err; + err = ret ? ret : err; goto restore_flag; } @@ -1643,7 +2144,8 @@ static int f2fs_disable_checkpoint(struct f2fs_sb_info *sbi) goto restore_flag; } - down_write(&sbi->gc_lock); +skip_gc: + f2fs_down_write(&sbi->gc_lock); cpc.reason = CP_PAUSE; set_sbi_flag(sbi, SBI_CP_DISABLED); err = f2fs_write_checkpoint(sbi, &cpc); @@ -1655,22 +2157,37 @@ static int f2fs_disable_checkpoint(struct f2fs_sb_info *sbi) spin_unlock(&sbi->stat_lock); out_unlock: - up_write(&sbi->gc_lock); + f2fs_up_write(&sbi->gc_lock); restore_flag: - sbi->sb->s_flags = s_flags; /* Restore MS_RDONLY status */ + sbi->gc_mode = gc_mode; + sbi->sb->s_flags = s_flags; /* Restore SB_RDONLY status */ return err; } static void f2fs_enable_checkpoint(struct f2fs_sb_info *sbi) { - down_write(&sbi->gc_lock); + int retry = DEFAULT_RETRY_IO_COUNT; + + /* we should flush all the data to keep data consistency */ + do { + sync_inodes_sb(sbi->sb); + f2fs_io_schedule_timeout(DEFAULT_IO_TIMEOUT); + } while (get_pages(sbi, F2FS_DIRTY_DATA) && retry--); + + if (unlikely(retry < 0)) + f2fs_warn(sbi, "checkpoint=enable has some unwritten data."); + + f2fs_down_write(&sbi->gc_lock); f2fs_dirty_to_prefree(sbi); clear_sbi_flag(sbi, SBI_CP_DISABLED); set_sbi_flag(sbi, SBI_IS_DIRTY); - up_write(&sbi->gc_lock); + f2fs_up_write(&sbi->gc_lock); f2fs_sync_fs(sbi->sb, 1); + + /* Let's ensure there's no pending checkpoint anymore */ + f2fs_flush_ckpt_thread(sbi); } static int f2fs_remount(struct super_block *sb, int *flags, char *data) @@ -1679,12 +2196,18 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) struct f2fs_mount_info org_mount_opt; unsigned long old_sb_flags; int err; - bool need_restart_gc = false; - bool need_stop_gc = false; + bool need_restart_gc = false, need_stop_gc = false; + bool need_restart_ckpt = false, need_stop_ckpt = false; + bool need_restart_flush = false, need_stop_flush = false; + bool need_restart_discard = false, need_stop_discard = false; bool no_extent_cache = !test_opt(sbi, EXTENT_CACHE); - bool disable_checkpoint = test_opt(sbi, DISABLE_CHECKPOINT); + bool enable_checkpoint = !test_opt(sbi, DISABLE_CHECKPOINT); bool no_io_align = !F2FS_IO_ALIGNED(sbi); - bool checkpoint_changed; + bool no_atgc = !test_opt(sbi, ATGC); + bool no_discard = !test_opt(sbi, DISCARD); + bool no_compress_cache = !test_opt(sbi, COMPRESS_CACHE); + bool block_unit_discard = f2fs_block_unit_discard(sbi); + struct discard_cmd_control *dcc; #ifdef CONFIG_QUOTA int i, j; #endif @@ -1705,7 +2228,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) GFP_KERNEL); if (!org_mount_opt.s_qf_names[i]) { for (j = 0; j < i; j++) - kvfree(org_mount_opt.s_qf_names[j]); + kfree(org_mount_opt.s_qf_names[j]); return -ENOMEM; } } else { @@ -1726,11 +2249,9 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) default_options(sbi); /* parse mount options */ - err = parse_options(sb, data); + err = parse_options(sb, data, true); if (err) goto restore_opts; - checkpoint_changed = - disable_checkpoint != test_opt(sbi, DISABLE_CHECKPOINT); /* * Previous and new state of filesystem is RO, @@ -1739,6 +2260,11 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) if (f2fs_readonly(sb) && (*flags & SB_RDONLY)) goto skip; + if (f2fs_sb_has_readonly(sbi) && !(*flags & SB_RDONLY)) { + err = -EROFS; + goto restore_opts; + } + #ifdef CONFIG_QUOTA if (!f2fs_readonly(sb) && (*flags & SB_RDONLY)) { err = dquot_suspend(sb, -1); @@ -1756,6 +2282,13 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) } } #endif + /* disallow enable atgc dynamically */ + if (no_atgc == !!test_opt(sbi, ATGC)) { + err = -EINVAL; + f2fs_warn(sbi, "switch atgc option is not allowed"); + goto restore_opts; + } + /* disallow enable/disable extent_cache dynamically */ if (no_extent_cache == !!test_opt(sbi, EXTENT_CACHE)) { err = -EINVAL; @@ -1769,6 +2302,18 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) goto restore_opts; } + if (no_compress_cache == !!test_opt(sbi, COMPRESS_CACHE)) { + err = -EINVAL; + f2fs_warn(sbi, "switch compress_cache option is not allowed"); + goto restore_opts; + } + + if (block_unit_discard != f2fs_block_unit_discard(sbi)) { + err = -EINVAL; + f2fs_warn(sbi, "switch discard_unit option is not allowed"); + goto restore_opts; + } + if ((*flags & SB_RDONLY) && test_opt(sbi, DISABLE_CHECKPOINT)) { err = -EINVAL; f2fs_warn(sbi, "disabling checkpoint not compatible with read-only"); @@ -1780,7 +2325,9 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) * or if background_gc = off is passed in mount * option. Also sync the filesystem. */ - if ((*flags & SB_RDONLY) || !test_opt(sbi, BG_GC)) { + if ((*flags & SB_RDONLY) || + (F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_OFF && + !test_opt(sbi, GC_MERGE))) { if (sbi->gc_thread) { f2fs_stop_gc_thread(sbi); need_restart_gc = true; @@ -1792,9 +2339,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) need_stop_gc = true; } - if (*flags & SB_RDONLY || - F2FS_OPTION(sbi).whint_mode != org_mount_opt.whint_mode) { - writeback_inodes_sb(sb, WB_REASON_SYNC); + if (*flags & SB_RDONLY) { sync_inodes_sb(sb); set_sbi_flag(sbi, SBI_IS_DIRTY); @@ -1803,14 +2348,22 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) clear_sbi_flag(sbi, SBI_IS_CLOSE); } - if (checkpoint_changed) { - if (test_opt(sbi, DISABLE_CHECKPOINT)) { - err = f2fs_disable_checkpoint(sbi); - if (err) - goto restore_gc; - } else { - f2fs_enable_checkpoint(sbi); + if ((*flags & SB_RDONLY) || test_opt(sbi, DISABLE_CHECKPOINT) || + !test_opt(sbi, MERGE_CHECKPOINT)) { + f2fs_stop_ckpt_thread(sbi); + need_restart_ckpt = true; + } else { + /* Flush if the prevous checkpoint, if exists. */ + f2fs_flush_ckpt_thread(sbi); + + err = f2fs_start_ckpt_thread(sbi); + if (err) { + f2fs_err(sbi, + "Failed to start F2FS issue_checkpoint_thread (%d)", + err); + goto restore_gc; } + need_stop_ckpt = true; } /* @@ -1820,24 +2373,75 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) if ((*flags & SB_RDONLY) || !test_opt(sbi, FLUSH_MERGE)) { clear_opt(sbi, FLUSH_MERGE); f2fs_destroy_flush_cmd_control(sbi, false); + need_restart_flush = true; } else { err = f2fs_create_flush_cmd_control(sbi); if (err) - goto restore_gc; + goto restore_ckpt; + need_stop_flush = true; } + + if (no_discard == !!test_opt(sbi, DISCARD)) { + if (test_opt(sbi, DISCARD)) { + err = f2fs_start_discard_thread(sbi); + if (err) + goto restore_flush; + need_stop_discard = true; + } else { + dcc = SM_I(sbi)->dcc_info; + f2fs_stop_discard_thread(sbi); + if (atomic_read(&dcc->discard_cmd_cnt)) + f2fs_issue_discard_timeout(sbi); + need_restart_discard = true; + } + } + + if (enable_checkpoint == !!test_opt(sbi, DISABLE_CHECKPOINT)) { + if (test_opt(sbi, DISABLE_CHECKPOINT)) { + err = f2fs_disable_checkpoint(sbi); + if (err) + goto restore_discard; + } else { + f2fs_enable_checkpoint(sbi); + } + } + skip: #ifdef CONFIG_QUOTA /* Release old quota file names */ for (i = 0; i < MAXQUOTAS; i++) - kvfree(org_mount_opt.s_qf_names[i]); + kfree(org_mount_opt.s_qf_names[i]); #endif /* Update the POSIXACL Flag */ sb->s_flags = (sb->s_flags & ~SB_POSIXACL) | (test_opt(sbi, POSIX_ACL) ? SB_POSIXACL : 0); limit_reserve_root(sbi); + adjust_unusable_cap_perc(sbi); *flags = (*flags & ~SB_LAZYTIME) | (sb->s_flags & SB_LAZYTIME); return 0; +restore_discard: + if (need_restart_discard) { + if (f2fs_start_discard_thread(sbi)) + f2fs_warn(sbi, "discard has been stopped"); + } else if (need_stop_discard) { + f2fs_stop_discard_thread(sbi); + } +restore_flush: + if (need_restart_flush) { + if (f2fs_create_flush_cmd_control(sbi)) + f2fs_warn(sbi, "background flush thread has stopped"); + } else if (need_stop_flush) { + clear_opt(sbi, FLUSH_MERGE); + f2fs_destroy_flush_cmd_control(sbi, false); + } +restore_ckpt: + if (need_restart_ckpt) { + if (f2fs_start_ckpt_thread(sbi)) + f2fs_warn(sbi, "background ckpt thread has stopped"); + } else if (need_stop_ckpt) { + f2fs_stop_ckpt_thread(sbi); + } restore_gc: if (need_restart_gc) { if (f2fs_start_gc_thread(sbi)) @@ -1849,7 +2453,7 @@ restore_opts: #ifdef CONFIG_QUOTA F2FS_OPTION(sbi).s_jquota_fmt = org_mount_opt.s_jquota_fmt; for (i = 0; i < MAXQUOTAS; i++) { - kvfree(F2FS_OPTION(sbi).s_qf_names[i]); + kfree(F2FS_OPTION(sbi).s_qf_names[i]); F2FS_OPTION(sbi).s_qf_names[i] = org_mount_opt.s_qf_names[i]; } #endif @@ -1871,7 +2475,6 @@ static ssize_t f2fs_quota_read(struct super_block *sb, int type, char *data, size_t toread; loff_t i_size = i_size_read(inode); struct page *page; - char *kaddr; if (off > i_size) return 0; @@ -1885,7 +2488,7 @@ repeat: page = read_cache_page_gfp(mapping, blkidx, GFP_NOFS); if (IS_ERR(page)) { if (PTR_ERR(page) == -ENOMEM) { - congestion_wait(BLK_RW_ASYNC, HZ/50); + memalloc_retry_wait(GFP_NOFS); goto repeat; } set_sbi_flag(F2FS_SB(sb), SBI_QUOTA_NEED_REPAIR); @@ -1904,9 +2507,7 @@ repeat: return -EIO; } - kaddr = kmap_atomic(page); - memcpy(data, kaddr + offset, tocopy); - kunmap_atomic(kaddr); + memcpy_from_page(data, page, offset, tocopy); f2fs_put_page(page, 1); offset = 0; @@ -1927,7 +2528,7 @@ static ssize_t f2fs_quota_write(struct super_block *sb, int type, int offset = off & (sb->s_blocksize - 1); size_t towrite = len; struct page *page; - char *kaddr; + void *fsdata = NULL; int err = 0; int tocopy; @@ -1935,24 +2536,21 @@ static ssize_t f2fs_quota_write(struct super_block *sb, int type, tocopy = min_t(unsigned long, sb->s_blocksize - offset, towrite); retry: - err = a_ops->write_begin(NULL, mapping, off, tocopy, 0, - &page, NULL); + err = a_ops->write_begin(NULL, mapping, off, tocopy, + &page, &fsdata); if (unlikely(err)) { if (err == -ENOMEM) { - congestion_wait(BLK_RW_ASYNC, HZ/50); + f2fs_io_schedule_timeout(DEFAULT_IO_TIMEOUT); goto retry; } set_sbi_flag(F2FS_SB(sb), SBI_QUOTA_NEED_REPAIR); break; } - kaddr = kmap_atomic(page); - memcpy(kaddr + offset, data, tocopy); - kunmap_atomic(kaddr); - flush_dcache_page(page); + memcpy_to_page(page, offset, data, tocopy); a_ops->write_end(NULL, mapping, off, tocopy, tocopy, - page, NULL); + page, fsdata); offset = 0; towrite -= tocopy; off += tocopy; @@ -1967,6 +2565,16 @@ retry: return len - towrite; } +int f2fs_dquot_initialize(struct inode *inode) +{ + if (time_to_inject(F2FS_I_SB(inode), FAULT_DQUOT_INIT)) { + f2fs_show_injection_info(F2FS_I_SB(inode), FAULT_DQUOT_INIT); + return -ESRCH; + } + + return dquot_initialize(inode); +} + static struct dquot **f2fs_get_dquots(struct inode *inode) { return F2FS_I(inode)->i_dquot; @@ -2080,64 +2688,78 @@ static int f2fs_enable_quotas(struct super_block *sb) return 0; } -int f2fs_quota_sync(struct super_block *sb, int type) +static int f2fs_quota_sync_file(struct f2fs_sb_info *sbi, int type) { - struct f2fs_sb_info *sbi = F2FS_SB(sb); - struct quota_info *dqopt = sb_dqopt(sb); - int cnt; - int ret; + struct quota_info *dqopt = sb_dqopt(sbi->sb); + struct address_space *mapping = dqopt->files[type]->i_mapping; + int ret = 0; - /* - * do_quotactl - * f2fs_quota_sync - * down_read(quota_sem) - * dquot_writeback_dquots() - * f2fs_dquot_commit - * block_operation - * down_read(quota_sem) - */ - f2fs_lock_op(sbi); + ret = dquot_writeback_dquots(sbi->sb, type); + if (ret) + goto out; - down_read(&sbi->quota_sem); - ret = dquot_writeback_dquots(sb, type); + ret = filemap_fdatawrite(mapping); if (ret) goto out; + /* if we are using journalled quota */ + if (is_journalled_quota(sbi)) + goto out; + + ret = filemap_fdatawait(mapping); + + truncate_inode_pages(&dqopt->files[type]->i_data, 0); +out: + if (ret) + set_sbi_flag(sbi, SBI_QUOTA_NEED_REPAIR); + return ret; +} + +int f2fs_quota_sync(struct super_block *sb, int type) +{ + struct f2fs_sb_info *sbi = F2FS_SB(sb); + struct quota_info *dqopt = sb_dqopt(sb); + int cnt; + int ret = 0; + /* * Now when everything is written we can discard the pagecache so * that userspace sees the changes. */ for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - struct address_space *mapping; if (type != -1 && cnt != type) continue; + if (!sb_has_quota_active(sb, cnt)) continue; - mapping = dqopt->files[cnt]->i_mapping; + if (!f2fs_sb_has_quota_ino(sbi)) + inode_lock(dqopt->files[cnt]); - ret = filemap_fdatawrite(mapping); - if (ret) - goto out; + /* + * do_quotactl + * f2fs_quota_sync + * f2fs_down_read(quota_sem) + * dquot_writeback_dquots() + * f2fs_dquot_commit + * block_operation + * f2fs_down_read(quota_sem) + */ + f2fs_lock_op(sbi); + f2fs_down_read(&sbi->quota_sem); - /* if we are using journalled quota */ - if (is_journalled_quota(sbi)) - continue; + ret = f2fs_quota_sync_file(sbi, cnt); - ret = filemap_fdatawait(mapping); - if (ret) - set_sbi_flag(F2FS_SB(sb), SBI_QUOTA_NEED_REPAIR); + f2fs_up_read(&sbi->quota_sem); + f2fs_unlock_op(sbi); + + if (!f2fs_sb_has_quota_ino(sbi)) + inode_unlock(dqopt->files[cnt]); - inode_lock(dqopt->files[cnt]); - truncate_inode_pages(&dqopt->files[cnt]->i_data, 0); - inode_unlock(dqopt->files[cnt]); + if (ret) + break; } -out: - if (ret) - set_sbi_flag(F2FS_SB(sb), SBI_QUOTA_NEED_REPAIR); - up_read(&sbi->quota_sem); - f2fs_unlock_op(sbi); return ret; } @@ -2255,11 +2877,11 @@ static int f2fs_dquot_commit(struct dquot *dquot) struct f2fs_sb_info *sbi = F2FS_SB(dquot->dq_sb); int ret; - down_read_nested(&sbi->quota_sem, SINGLE_DEPTH_NESTING); + f2fs_down_read_nested(&sbi->quota_sem, SINGLE_DEPTH_NESTING); ret = dquot_commit(dquot); if (ret < 0) set_sbi_flag(sbi, SBI_QUOTA_NEED_REPAIR); - up_read(&sbi->quota_sem); + f2fs_up_read(&sbi->quota_sem); return ret; } @@ -2268,11 +2890,11 @@ static int f2fs_dquot_acquire(struct dquot *dquot) struct f2fs_sb_info *sbi = F2FS_SB(dquot->dq_sb); int ret; - down_read(&sbi->quota_sem); + f2fs_down_read(&sbi->quota_sem); ret = dquot_acquire(dquot); if (ret < 0) set_sbi_flag(sbi, SBI_QUOTA_NEED_REPAIR); - up_read(&sbi->quota_sem); + f2fs_up_read(&sbi->quota_sem); return ret; } @@ -2339,6 +2961,11 @@ static const struct quotactl_ops f2fs_quotactl_ops = { .get_nextdqblk = dquot_get_next_dqblk, }; #else +int f2fs_dquot_initialize(struct inode *inode) +{ + return 0; +} + int f2fs_quota_sync(struct super_block *sb, int type) { return 0; @@ -2398,9 +3025,9 @@ static int f2fs_set_context(struct inode *inode, const void *ctx, size_t len, ctx, len, fs_data, XATTR_CREATE); } -static bool f2fs_dummy_context(struct inode *inode) +static const union fscrypt_policy *f2fs_get_dummy_policy(struct super_block *sb) { - return DUMMY_ENCRYPTION_ENABLED(F2FS_I_SB(inode)); + return F2FS_OPTION(F2FS_SB(sb)).dummy_enc_policy.policy; } static bool f2fs_has_stable_inodes(struct super_block *sb) @@ -2415,15 +3042,35 @@ static void f2fs_get_ino_and_lblk_bits(struct super_block *sb, *lblk_bits_ret = 8 * sizeof(block_t); } +static struct block_device **f2fs_get_devices(struct super_block *sb, + unsigned int *num_devs) +{ + struct f2fs_sb_info *sbi = F2FS_SB(sb); + struct block_device **devs; + int i; + + if (!f2fs_is_multi_device(sbi)) + return NULL; + + devs = kmalloc_array(sbi->s_ndevs, sizeof(*devs), GFP_KERNEL); + if (!devs) + return ERR_PTR(-ENOMEM); + + for (i = 0; i < sbi->s_ndevs; i++) + devs[i] = FDEV(i).bdev; + *num_devs = sbi->s_ndevs; + return devs; +} + static const struct fscrypt_operations f2fs_cryptops = { .key_prefix = "f2fs:", .get_context = f2fs_get_context, .set_context = f2fs_set_context, - .dummy_context = f2fs_dummy_context, + .get_dummy_policy = f2fs_get_dummy_policy, .empty_dir = f2fs_empty_dir, - .max_namelen = F2FS_NAME_LEN, .has_stable_inodes = f2fs_has_stable_inodes, .get_ino_and_lblk_bits = f2fs_get_ino_and_lblk_bits, + .get_devices = f2fs_get_devices, }; #endif @@ -2472,10 +3119,10 @@ static const struct export_operations f2fs_export_ops = { .get_parent = f2fs_get_parent, }; -static loff_t max_file_blocks(void) +loff_t max_file_blocks(struct inode *inode) { loff_t result = 0; - loff_t leaf_count = DEF_ADDRS_PER_BLOCK; + loff_t leaf_count; /* * note: previously, result is equal to (DEF_ADDRS_PER_INODE - @@ -2484,6 +3131,11 @@ static loff_t max_file_blocks(void) * result as zero. */ + if (inode && f2fs_compressed_file(inode)) + leaf_count = ADDRS_PER_BLOCK(inode); + else + leaf_count = DEF_ADDRS_PER_BLOCK; + /* two direct node blocks */ result += (leaf_count * 2); @@ -2574,10 +3226,8 @@ static inline bool sanity_check_area_boundary(struct f2fs_sb_info *sbi, } if (main_end_blkaddr > seg_end_blkaddr) { - f2fs_info(sbi, "Wrong MAIN_AREA boundary, start(%u) end(%u) block(%u)", - main_blkaddr, - segment0_blkaddr + - (segment_count << log_blocks_per_seg), + f2fs_info(sbi, "Wrong MAIN_AREA boundary, start(%u) end(%llu) block(%u)", + main_blkaddr, seg_end_blkaddr, segment_count_main << log_blocks_per_seg); return true; } else if (main_end_blkaddr < seg_end_blkaddr) { @@ -2595,10 +3245,8 @@ static inline bool sanity_check_area_boundary(struct f2fs_sb_info *sbi, err = __f2fs_commit_super(bh, NULL); res = err ? "failed" : "done"; } - f2fs_info(sbi, "Fix alignment : %s, start(%u) end(%u) block(%u)", - res, main_blkaddr, - segment0_blkaddr + - (segment_count << log_blocks_per_seg), + f2fs_info(sbi, "Fix alignment : %s, start(%u) end(%llu) block(%u)", + res, main_blkaddr, seg_end_blkaddr, segment_count_main << log_blocks_per_seg); if (err) return true; @@ -2609,11 +3257,10 @@ static inline bool sanity_check_area_boundary(struct f2fs_sb_info *sbi, static int sanity_check_raw_super(struct f2fs_sb_info *sbi, struct buffer_head *bh) { - block_t segment_count, segs_per_sec, secs_per_zone; + block_t segment_count, segs_per_sec, secs_per_zone, segment_count_main; block_t total_sections, blocks_per_seg; struct f2fs_super_block *raw_super = (struct f2fs_super_block *) (bh->b_data + F2FS_SUPER_OFFSET); - unsigned int blocksize; size_t crc_offset = 0; __u32 crc = 0; @@ -2639,18 +3286,11 @@ static int sanity_check_raw_super(struct f2fs_sb_info *sbi, } } - /* Currently, support only 4KB page cache size */ - if (F2FS_BLKSIZE != PAGE_SIZE) { - f2fs_info(sbi, "Invalid page_cache_size (%lu), supports only 4KB", - PAGE_SIZE); - return -EFSCORRUPTED; - } - /* Currently, support only 4KB block size */ - blocksize = 1 << le32_to_cpu(raw_super->log_blocksize); - if (blocksize != F2FS_BLKSIZE) { - f2fs_info(sbi, "Invalid blocksize (%u), supports only 4KB", - blocksize); + if (le32_to_cpu(raw_super->log_blocksize) != F2FS_BLKSIZE_BITS) { + f2fs_info(sbi, "Invalid log_blocksize (%u), supports only %u", + le32_to_cpu(raw_super->log_blocksize), + F2FS_BLKSIZE_BITS); return -EFSCORRUPTED; } @@ -2680,6 +3320,7 @@ static int sanity_check_raw_super(struct f2fs_sb_info *sbi, } segment_count = le32_to_cpu(raw_super->segment_count); + segment_count_main = le32_to_cpu(raw_super->segment_count_main); segs_per_sec = le32_to_cpu(raw_super->segs_per_sec); secs_per_zone = le32_to_cpu(raw_super->secs_per_zone); total_sections = le32_to_cpu(raw_super->section_count); @@ -2693,14 +3334,19 @@ static int sanity_check_raw_super(struct f2fs_sb_info *sbi, return -EFSCORRUPTED; } - if (total_sections > segment_count || - total_sections < F2FS_MIN_SEGMENTS || + if (total_sections > segment_count_main || total_sections < 1 || segs_per_sec > segment_count || !segs_per_sec) { f2fs_info(sbi, "Invalid segment/section count (%u, %u x %u)", segment_count, total_sections, segs_per_sec); return -EFSCORRUPTED; } + if (segment_count_main != total_sections * segs_per_sec) { + f2fs_info(sbi, "Invalid segment/section count (%u != %u * %u)", + segment_count_main, total_sections, segs_per_sec); + return -EFSCORRUPTED; + } + if ((segment_count / segs_per_sec) < total_sections) { f2fs_info(sbi, "Small segment_count (%u < %u * %u)", segment_count, segs_per_sec, total_sections); @@ -2726,6 +3372,12 @@ static int sanity_check_raw_super(struct f2fs_sb_info *sbi, segment_count, dev_seg_count); return -EFSCORRUPTED; } + } else { + if (__F2FS_HAS_FEATURE(raw_super, F2FS_FEATURE_BLKZONED) && + !bdev_is_zoned(sbi->sb->s_bdev)) { + f2fs_info(sbi, "Zoned block device path is missing"); + return -EFSCORRUPTED; + } } if (secs_per_zone > total_sections || !secs_per_zone) { @@ -2744,11 +3396,13 @@ static int sanity_check_raw_super(struct f2fs_sb_info *sbi, return -EFSCORRUPTED; } - if (le32_to_cpu(raw_super->cp_payload) > - (blocks_per_seg - F2FS_CP_PACKS)) { - f2fs_info(sbi, "Insane cp_payload (%u > %u)", + if (le32_to_cpu(raw_super->cp_payload) >= + (blocks_per_seg - F2FS_CP_PACKS - + NR_CURSEG_PERSIST_TYPE)) { + f2fs_info(sbi, "Insane cp_payload (%u >= %u)", le32_to_cpu(raw_super->cp_payload), - blocks_per_seg - F2FS_CP_PACKS); + blocks_per_seg - F2FS_CP_PACKS - + NR_CURSEG_PERSIST_TYPE); return -EFSCORRUPTED; } @@ -2784,6 +3438,7 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi) unsigned int cp_pack_start_sum, cp_payload; block_t user_block_count, valid_user_blocks; block_t avail_node_count, valid_node_count; + unsigned int nat_blocks, nat_bits_bytes, nat_bits_blocks; int i, j; total = le32_to_cpu(raw_super->segment_count); @@ -2801,14 +3456,15 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi) ovp_segments = le32_to_cpu(ckpt->overprov_segment_count); reserved_segments = le32_to_cpu(ckpt->rsvd_segment_count); - if (unlikely(fsmeta < F2FS_MIN_SEGMENTS || + if (!f2fs_sb_has_readonly(sbi) && + unlikely(fsmeta < F2FS_MIN_META_SEGMENTS || ovp_segments == 0 || reserved_segments == 0)) { f2fs_err(sbi, "Wrong layout: check mkfs.f2fs version"); return 1; } - user_block_count = le64_to_cpu(ckpt->user_block_count); - segment_count_main = le32_to_cpu(raw_super->segment_count_main); + segment_count_main = le32_to_cpu(raw_super->segment_count_main) + + (f2fs_sb_has_readonly(sbi) ? 1 : 0); log_blocks_per_seg = le32_to_cpu(raw_super->log_blocks_per_seg); if (!user_block_count || user_block_count >= segment_count_main << log_blocks_per_seg) { @@ -2839,6 +3495,10 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi) if (le32_to_cpu(ckpt->cur_node_segno[i]) >= main_segs || le16_to_cpu(ckpt->cur_node_blkoff[i]) >= blocks_per_seg) return 1; + + if (f2fs_sb_has_readonly(sbi)) + goto check_data; + for (j = i + 1; j < NR_CURSEG_NODE_TYPE; j++) { if (le32_to_cpu(ckpt->cur_node_segno[i]) == le32_to_cpu(ckpt->cur_node_segno[j])) { @@ -2849,10 +3509,15 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi) } } } +check_data: for (i = 0; i < NR_CURSEG_DATA_TYPE; i++) { if (le32_to_cpu(ckpt->cur_data_segno[i]) >= main_segs || le16_to_cpu(ckpt->cur_data_blkoff[i]) >= blocks_per_seg) return 1; + + if (f2fs_sb_has_readonly(sbi)) + goto skip_cross; + for (j = i + 1; j < NR_CURSEG_DATA_TYPE; j++) { if (le32_to_cpu(ckpt->cur_data_segno[i]) == le32_to_cpu(ckpt->cur_data_segno[j])) { @@ -2874,7 +3539,7 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi) } } } - +skip_cross: sit_bitmap_size = le32_to_cpu(ckpt->sit_ver_bitmap_bytesize); nat_bitmap_size = le32_to_cpu(ckpt->nat_ver_bitmap_bytesize); @@ -2889,7 +3554,7 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi) cp_payload = __cp_payload(sbi); if (cp_pack_start_sum < cp_payload + 1 || cp_pack_start_sum > blocks_per_seg - 1 - - NR_CURSEG_TYPE) { + NR_CURSEG_PERSIST_TYPE) { f2fs_err(sbi, "Wrong cp_pack_start_sum: %u", cp_pack_start_sum); return 1; @@ -2904,6 +3569,17 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi) return 1; } + nat_blocks = nat_segs << log_blocks_per_seg; + nat_bits_bytes = nat_blocks / BITS_PER_BYTE; + nat_bits_blocks = F2FS_BLK_ALIGN((nat_bits_bytes << 1) + 8); + if (__is_set_ckpt_flags(ckpt, CP_NAT_BITS_FLAG) && + (cp_payload + F2FS_CP_PACKS + + NR_CURSEG_PERSIST_TYPE + nat_bits_blocks >= blocks_per_seg)) { + f2fs_warn(sbi, "Insane cp_payload: %u, nat_bits_blocks: %u)", + cp_payload, nat_bits_blocks); + return 1; + } + if (unlikely(f2fs_cp_error(sbi))) { f2fs_err(sbi, "A bug case: need to run fsck"); return 1; @@ -2928,14 +3604,20 @@ static void init_sb_info(struct f2fs_sb_info *sbi) sbi->total_node_count = (le32_to_cpu(raw_super->segment_count_nat) / 2) * sbi->blocks_per_seg * NAT_ENTRY_PER_BLOCK; - sbi->root_ino_num = le32_to_cpu(raw_super->root_ino); - sbi->node_ino_num = le32_to_cpu(raw_super->node_ino); - sbi->meta_ino_num = le32_to_cpu(raw_super->meta_ino); + F2FS_ROOT_INO(sbi) = le32_to_cpu(raw_super->root_ino); + F2FS_NODE_INO(sbi) = le32_to_cpu(raw_super->node_ino); + F2FS_META_INO(sbi) = le32_to_cpu(raw_super->meta_ino); sbi->cur_victim_sec = NULL_SECNO; + sbi->gc_mode = GC_NORMAL; sbi->next_victim_seg[BG_GC] = NULL_SEGNO; sbi->next_victim_seg[FG_GC] = NULL_SEGNO; sbi->max_victim_search = DEF_MAX_VICTIM_SEARCH; sbi->migration_granularity = sbi->segs_per_sec; + sbi->seq_file_ra_mul = MIN_RA_MUL; + sbi->max_fragment_chunk = DEF_FRAGMENT_SIZE; + sbi->max_fragment_hole = DEF_FRAGMENT_SIZE; + spin_lock_init(&sbi->gc_urgent_high_lock); + atomic64_set(&sbi->current_atomic_write, 0); sbi->dir_level = DEF_DIR_LEVEL; sbi->interval_time[CP_TIME] = DEF_CP_INTERVAL; @@ -2955,14 +3637,14 @@ static void init_sb_info(struct f2fs_sb_info *sbi) INIT_LIST_HEAD(&sbi->s_list); mutex_init(&sbi->umount_mutex); - init_rwsem(&sbi->io_order_lock); + init_f2fs_rwsem(&sbi->io_order_lock); spin_lock_init(&sbi->cp_lock); sbi->dirty_device = 0; spin_lock_init(&sbi->dev_lock); - init_rwsem(&sbi->sb_lock); - init_rwsem(&sbi->pin_sem); + init_f2fs_rwsem(&sbi->sb_lock); + init_f2fs_rwsem(&sbi->pin_sem); } static int init_percpu_info(struct f2fs_sb_info *sbi) @@ -2973,60 +3655,96 @@ static int init_percpu_info(struct f2fs_sb_info *sbi) if (err) return err; + err = percpu_counter_init(&sbi->rf_node_block_count, 0, GFP_KERNEL); + if (err) + goto err_valid_block; + err = percpu_counter_init(&sbi->total_valid_inode_count, 0, GFP_KERNEL); if (err) - percpu_counter_destroy(&sbi->alloc_valid_block_count); + goto err_node_block; + return 0; +err_node_block: + percpu_counter_destroy(&sbi->rf_node_block_count); +err_valid_block: + percpu_counter_destroy(&sbi->alloc_valid_block_count); return err; } #ifdef CONFIG_BLK_DEV_ZONED + +struct f2fs_report_zones_args { + struct f2fs_sb_info *sbi; + struct f2fs_dev_info *dev; +}; + static int f2fs_report_zone_cb(struct blk_zone *zone, unsigned int idx, - void *data) + void *data) { - struct f2fs_dev_info *dev = data; + struct f2fs_report_zones_args *rz_args = data; + block_t unusable_blocks = (zone->len - zone->capacity) >> + F2FS_LOG_SECTORS_PER_BLOCK; + + if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL) + return 0; - if (zone->type != BLK_ZONE_TYPE_CONVENTIONAL) - set_bit(idx, dev->blkz_seq); + set_bit(idx, rz_args->dev->blkz_seq); + if (!rz_args->sbi->unusable_blocks_per_sec) { + rz_args->sbi->unusable_blocks_per_sec = unusable_blocks; + return 0; + } + if (rz_args->sbi->unusable_blocks_per_sec != unusable_blocks) { + f2fs_err(rz_args->sbi, "F2FS supports single zone capacity\n"); + return -EINVAL; + } return 0; } static int init_blkz_info(struct f2fs_sb_info *sbi, int devi) { struct block_device *bdev = FDEV(devi).bdev; - sector_t nr_sectors = bdev->bd_part->nr_sects; + sector_t nr_sectors = bdev_nr_sectors(bdev); + struct f2fs_report_zones_args rep_zone_arg; + u64 zone_sectors; int ret; if (!f2fs_sb_has_blkzoned(sbi)) return 0; + zone_sectors = bdev_zone_sectors(bdev); + if (!is_power_of_2(zone_sectors)) { + f2fs_err(sbi, "F2FS does not support non power of 2 zone sizes\n"); + return -EINVAL; + } + if (sbi->blocks_per_blkz && sbi->blocks_per_blkz != - SECTOR_TO_BLOCK(bdev_zone_sectors(bdev))) + SECTOR_TO_BLOCK(zone_sectors)) return -EINVAL; - sbi->blocks_per_blkz = SECTOR_TO_BLOCK(bdev_zone_sectors(bdev)); + sbi->blocks_per_blkz = SECTOR_TO_BLOCK(zone_sectors); if (sbi->log_blocks_per_blkz && sbi->log_blocks_per_blkz != __ilog2_u32(sbi->blocks_per_blkz)) return -EINVAL; sbi->log_blocks_per_blkz = __ilog2_u32(sbi->blocks_per_blkz); FDEV(devi).nr_blkz = SECTOR_TO_BLOCK(nr_sectors) >> sbi->log_blocks_per_blkz; - if (nr_sectors & (bdev_zone_sectors(bdev) - 1)) + if (nr_sectors & (zone_sectors - 1)) FDEV(devi).nr_blkz++; - FDEV(devi).blkz_seq = f2fs_kzalloc(sbi, + FDEV(devi).blkz_seq = f2fs_kvzalloc(sbi, BITS_TO_LONGS(FDEV(devi).nr_blkz) * sizeof(unsigned long), GFP_KERNEL); if (!FDEV(devi).blkz_seq) return -ENOMEM; - /* Get block zones type */ + rep_zone_arg.sbi = sbi; + rep_zone_arg.dev = &FDEV(devi); + ret = blkdev_report_zones(bdev, 0, BLK_ALL_ZONES, f2fs_report_zone_cb, - &FDEV(devi)); + &rep_zone_arg); if (ret < 0) return ret; - return 0; } #endif @@ -3082,7 +3800,7 @@ static int read_raw_super_block(struct f2fs_sb_info *sbi, /* No valid superblock */ if (!*raw_super) - kvfree(super); + kfree(super); else err = 0; @@ -3128,10 +3846,73 @@ int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover) return err; } +void f2fs_handle_stop(struct f2fs_sb_info *sbi, unsigned char reason) +{ + struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi); + int err; + + f2fs_down_write(&sbi->sb_lock); + + if (raw_super->s_stop_reason[reason] < ((1 << BITS_PER_BYTE) - 1)) + raw_super->s_stop_reason[reason]++; + + err = f2fs_commit_super(sbi, false); + if (err) + f2fs_err(sbi, "f2fs_commit_super fails to record reason:%u err:%d", + reason, err); + f2fs_up_write(&sbi->sb_lock); +} + +static void f2fs_save_errors(struct f2fs_sb_info *sbi, unsigned char flag) +{ + spin_lock(&sbi->error_lock); + if (!test_bit(flag, (unsigned long *)sbi->errors)) { + set_bit(flag, (unsigned long *)sbi->errors); + sbi->error_dirty = true; + } + spin_unlock(&sbi->error_lock); +} + +static bool f2fs_update_errors(struct f2fs_sb_info *sbi) +{ + bool need_update = false; + + spin_lock(&sbi->error_lock); + if (sbi->error_dirty) { + memcpy(F2FS_RAW_SUPER(sbi)->s_errors, sbi->errors, + MAX_F2FS_ERRORS); + sbi->error_dirty = false; + need_update = true; + } + spin_unlock(&sbi->error_lock); + + return need_update; +} + +void f2fs_handle_error(struct f2fs_sb_info *sbi, unsigned char error) +{ + int err; + + f2fs_save_errors(sbi, error); + + f2fs_down_write(&sbi->sb_lock); + + if (!f2fs_update_errors(sbi)) + goto out_unlock; + + err = f2fs_commit_super(sbi, false); + if (err) + f2fs_err(sbi, "f2fs_commit_super fails to record errors:%u, err:%d", + error, err); +out_unlock: + f2fs_up_write(&sbi->sb_lock); +} + static int f2fs_scan_devices(struct f2fs_sb_info *sbi) { struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi); unsigned int max_devices = MAX_DEVICES; + unsigned int logical_blksize; int i; /* Initialize single device information */ @@ -3152,6 +3933,9 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi) if (!sbi->devs) return -ENOMEM; + logical_blksize = bdev_logical_block_size(sbi->sb->s_bdev); + sbi->aligned_blksize = true; + for (i = 0; i < max_devices; i++) { if (i > 0 && !RDEV(i).path[0]) @@ -3188,10 +3972,13 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi) /* to release errored devices */ sbi->s_ndevs = i + 1; + if (logical_blksize != bdev_logical_block_size(FDEV(i).bdev)) + sbi->aligned_blksize = false; + #ifdef CONFIG_BLK_DEV_ZONED if (bdev_zoned_model(FDEV(i).bdev) == BLK_ZONED_HM && !f2fs_sb_has_blkzoned(sbi)) { - f2fs_err(sbi, "Zoned block device feature not enabled\n"); + f2fs_err(sbi, "Zoned block device feature not enabled"); return -EINVAL; } if (bdev_zoned_model(FDEV(i).bdev) != BLK_ZONED_NONE) { @@ -3222,41 +4009,41 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi) static int f2fs_setup_casefold(struct f2fs_sb_info *sbi) { -#ifdef CONFIG_UNICODE - if (f2fs_sb_has_casefold(sbi) && !sbi->s_encoding) { +#if IS_ENABLED(CONFIG_UNICODE) + if (f2fs_sb_has_casefold(sbi) && !sbi->sb->s_encoding) { const struct f2fs_sb_encodings *encoding_info; struct unicode_map *encoding; __u16 encoding_flags; - if (f2fs_sb_has_encrypt(sbi)) { - f2fs_err(sbi, - "Can't mount with encoding and encryption"); - return -EINVAL; - } - - if (f2fs_sb_read_encoding(sbi->raw_super, &encoding_info, - &encoding_flags)) { + encoding_info = f2fs_sb_read_encoding(sbi->raw_super); + if (!encoding_info) { f2fs_err(sbi, "Encoding requested by superblock is unknown"); return -EINVAL; } + encoding_flags = le16_to_cpu(sbi->raw_super->s_encoding_flags); encoding = utf8_load(encoding_info->version); if (IS_ERR(encoding)) { f2fs_err(sbi, - "can't mount with superblock charset: %s-%s " + "can't mount with superblock charset: %s-%u.%u.%u " "not supported by the kernel. flags: 0x%x.", - encoding_info->name, encoding_info->version, + encoding_info->name, + unicode_major(encoding_info->version), + unicode_minor(encoding_info->version), + unicode_rev(encoding_info->version), encoding_flags); return PTR_ERR(encoding); } f2fs_info(sbi, "Using encoding defined by superblock: " - "%s-%s with flags 0x%hx", encoding_info->name, - encoding_info->version?:"\b", encoding_flags); + "%s-%u.%u.%u with flags 0x%hx", encoding_info->name, + unicode_major(encoding_info->version), + unicode_minor(encoding_info->version), + unicode_rev(encoding_info->version), + encoding_flags); - sbi->s_encoding = encoding; - sbi->s_encoding_flags = encoding_flags; - sbi->sb->s_d_op = &f2fs_dentry_ops; + sbi->sb->s_encoding = encoding; + sbi->sb->s_encoding_flags = encoding_flags; } #else if (f2fs_sb_has_casefold(sbi)) { @@ -3274,8 +4061,10 @@ static void f2fs_tuning_parameters(struct f2fs_sb_info *sbi) /* adjust parameters according to the volume size */ if (sm_i->main_segments <= SMALL_VOLUME_SEGMENTS) { F2FS_OPTION(sbi).alloc_mode = ALLOC_MODE_REUSE; - sm_i->dcc_info->discard_granularity = 1; - sm_i->ipu_policy = 1 << F2FS_IPU_FORCE; + if (f2fs_block_unit_discard(sbi)) + sm_i->dcc_info->discard_granularity = 1; + sm_i->ipu_policy = 1 << F2FS_IPU_FORCE | + 1 << F2FS_IPU_HONOR_OPU_WRITE; } sbi->readdir_ra = 1; @@ -3334,18 +4123,6 @@ try_onemore: sbi->s_chksum_seed = f2fs_chksum(sbi, ~0, raw_super->uuid, sizeof(raw_super->uuid)); - /* - * The BLKZONED feature indicates that the drive was formatted with - * zone alignment optimization. This is optional for host-aware - * devices, but mandatory for host-managed zoned block devices. - */ -#ifndef CONFIG_BLK_DEV_ZONED - if (f2fs_sb_has_blkzoned(sbi)) { - f2fs_err(sbi, "Zoned block device support is not enabled"); - err = -EOPNOTSUPP; - goto free_sb_buf; - } -#endif default_options(sbi); /* parse mount options */ options = kstrdup((const char *)data, GFP_KERNEL); @@ -3354,12 +4131,11 @@ try_onemore: goto free_sb_buf; } - err = parse_options(sb, options); + err = parse_options(sb, options, false); if (err) goto free_options; - sbi->max_file_blocks = max_file_blocks(); - sb->s_maxbytes = sbi->max_file_blocks << + sb->s_maxbytes = max_file_blocks(NULL) << le32_to_cpu(raw_super->log_blocksize); sb->s_max_links = F2FS_LINK_MAX; @@ -3398,55 +4174,33 @@ try_onemore: /* init f2fs-specific super block info */ sbi->valid_super_block = valid_super_block; - init_rwsem(&sbi->gc_lock); + init_f2fs_rwsem(&sbi->gc_lock); mutex_init(&sbi->writepages); - mutex_init(&sbi->cp_mutex); - mutex_init(&sbi->resize_mutex); - init_rwsem(&sbi->node_write); - init_rwsem(&sbi->node_change); + init_f2fs_rwsem(&sbi->cp_global_sem); + init_f2fs_rwsem(&sbi->node_write); + init_f2fs_rwsem(&sbi->node_change); /* disallow all the data/node/meta page writes */ set_sbi_flag(sbi, SBI_POR_DOING); spin_lock_init(&sbi->stat_lock); - /* init iostat info */ - spin_lock_init(&sbi->iostat_lock); - sbi->iostat_enable = false; - - for (i = 0; i < NR_PAGE_TYPE; i++) { - int n = (i == META) ? 1: NR_TEMP_TYPE; - int j; - - sbi->write_io[i] = - f2fs_kmalloc(sbi, - array_size(n, - sizeof(struct f2fs_bio_info)), - GFP_KERNEL); - if (!sbi->write_io[i]) { - err = -ENOMEM; - goto free_bio_info; - } - - for (j = HOT; j < n; j++) { - init_rwsem(&sbi->write_io[i][j].io_rwsem); - sbi->write_io[i][j].sbi = sbi; - sbi->write_io[i][j].bio = NULL; - spin_lock_init(&sbi->write_io[i][j].io_lock); - INIT_LIST_HEAD(&sbi->write_io[i][j].io_list); - INIT_LIST_HEAD(&sbi->write_io[i][j].bio_list); - init_rwsem(&sbi->write_io[i][j].bio_list_lock); - } - } + err = f2fs_init_write_merge_io(sbi); + if (err) + goto free_bio_info; - init_rwsem(&sbi->cp_rwsem); - init_rwsem(&sbi->quota_sem); + init_f2fs_rwsem(&sbi->cp_rwsem); + init_f2fs_rwsem(&sbi->quota_sem); init_waitqueue_head(&sbi->cp_wait); init_sb_info(sbi); - err = init_percpu_info(sbi); + err = f2fs_init_iostat(sbi); if (err) goto free_bio_info; + err = init_percpu_info(sbi); + if (err) + goto free_iostat; + if (F2FS_IO_ALIGNED(sbi)) { sbi->write_io_dummy = mempool_create_page_pool(2 * (F2FS_IO_SIZE(sbi) - 1), 0); @@ -3456,12 +4210,20 @@ try_onemore: } } + /* init per sbi slab cache */ + err = f2fs_init_xattr_caches(sbi); + if (err) + goto free_io_dummy; + err = f2fs_init_page_array_cache(sbi); + if (err) + goto free_xattr_cache; + /* get an inode for meta space */ sbi->meta_inode = f2fs_iget(sb, F2FS_META_INO(sbi)); if (IS_ERR(sbi->meta_inode)) { f2fs_err(sbi, "Failed to read F2FS meta data inode"); err = PTR_ERR(sbi->meta_inode); - goto free_io_dummy; + goto free_page_array_cache; } err = f2fs_get_valid_checkpoint(sbi); @@ -3493,6 +4255,9 @@ try_onemore: goto free_devices; } + spin_lock_init(&sbi->error_lock); + memcpy(sbi->errors, raw_super->s_errors, MAX_F2FS_ERRORS); + sbi->total_valid_node_count = le32_to_cpu(sbi->ckpt->valid_node_count); percpu_counter_set(&sbi->total_valid_inode_count, @@ -3504,6 +4269,7 @@ try_onemore: sbi->reserved_blocks = 0; sbi->current_reserved_blocks = 0; limit_reserve_root(sbi); + adjust_unusable_cap_perc(sbi); for (i = 0; i < NR_INODE_TYPE; i++) { INIT_LIST_HEAD(&sbi->inode_list[i]); @@ -3517,6 +4283,19 @@ try_onemore: f2fs_init_fsync_node_info(sbi); + /* setup checkpoint request control and start checkpoint issue thread */ + f2fs_init_ckpt_req_control(sbi); + if (!f2fs_readonly(sb) && !test_opt(sbi, DISABLE_CHECKPOINT) && + test_opt(sbi, MERGE_CHECKPOINT)) { + err = f2fs_start_ckpt_thread(sbi); + if (err) { + f2fs_err(sbi, + "Failed to start F2FS issue_checkpoint_thread (%d)", + err); + goto stop_ckpt_thread; + } + } + /* setup f2fs internal modules */ err = f2fs_build_segment_manager(sbi); if (err) { @@ -3531,11 +4310,12 @@ try_onemore: goto free_nm; } + err = adjust_reserved_segment(sbi); + if (err) + goto free_nm; + /* For write statistics */ - if (sb->s_bdev->bd_part) - sbi->sectors_written_start = - (u64)part_stat_read(sb->s_bdev->bd_part, - sectors[STAT_WRITE]); + sbi->sectors_written_start = f2fs_get_sectors_written(sbi); /* Read accumulated write IO statistics if exists */ seg_i = CURSEG_I(sbi, CURSEG_HOT_NODE); @@ -3577,10 +4357,14 @@ try_onemore: goto free_node_inode; } - err = f2fs_register_sysfs(sbi); + err = f2fs_init_compress_inode(sbi); if (err) goto free_root_inode; + err = f2fs_register_sysfs(sbi); + if (err) + goto free_compress_inode; + #ifdef CONFIG_QUOTA /* Enable quota usage during mount */ if (f2fs_sb_has_quota_ino(sbi) && !f2fs_readonly(sb)) { @@ -3589,7 +4373,7 @@ try_onemore: f2fs_err(sbi, "Cannot turn on quotas: error %d", err); } #endif - /* if there are nt orphan nodes free them */ + /* if there are any orphan inodes, free them */ err = f2fs_recover_orphan_inodes(sbi); if (err) goto free_meta; @@ -3598,16 +4382,23 @@ try_onemore: goto reset_checkpoint; /* recover fsynced data */ - if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) { + if (!test_opt(sbi, DISABLE_ROLL_FORWARD) && + !test_opt(sbi, NORECOVERY)) { /* * mount should be failed, when device has readonly mode, and * previous checkpoint was not done by clean system shutdown. */ if (f2fs_hw_is_readonly(sbi)) { if (!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) { - err = -EROFS; - f2fs_err(sbi, "Need to recover fsync data, but write access unavailable"); - goto free_meta; + err = f2fs_recover_fsync_data(sbi, true); + if (err > 0) { + err = -EROFS; + f2fs_err(sbi, "Need to recover fsync data, but " + "write access unavailable, please try " + "mount w/ disable_roll_forward or norecovery"); + } + if (err < 0) + goto free_meta; } f2fs_info(sbi, "write access unavailable, skipping recovery"); goto reset_checkpoint; @@ -3649,6 +4440,8 @@ try_onemore: } reset_checkpoint: + f2fs_init_inmem_curseg(sbi); + /* f2fs_recover_fsync_data() cleared this already */ clear_sbi_flag(sbi, SBI_POR_DOING); @@ -3664,7 +4457,8 @@ reset_checkpoint: * If filesystem is not mounted as read-only then * do start the gc_thread. */ - if (test_opt(sbi, BG_GC) && !f2fs_readonly(sb)) { + if ((F2FS_OPTION(sbi).bggc_mode != BGGC_MODE_OFF || + test_opt(sbi, GC_MERGE)) && !f2fs_readonly(sb)) { /* After POR, we can run background GC thread.*/ err = f2fs_start_gc_thread(sbi); if (err) @@ -3711,6 +4505,8 @@ free_meta: /* evict some inodes being cached by GC */ evict_inodes(sb); f2fs_unregister_sysfs(sbi); +free_compress_inode: + f2fs_destroy_compress_inode(sbi); free_root_inode: dput(sb->s_root); sb->s_root = NULL; @@ -3722,10 +4518,14 @@ free_node_inode: free_stats: f2fs_destroy_stats(sbi); free_nm: + /* stop discard thread before destroying node manager */ + f2fs_stop_discard_thread(sbi); f2fs_destroy_node_manager(sbi); free_sm: f2fs_destroy_segment_manager(sbi); f2fs_destroy_post_read_wq(sbi); +stop_ckpt_thread: + f2fs_stop_ckpt_thread(sbi); free_devices: destroy_device_list(sbi); kvfree(sbi->ckpt); @@ -3733,29 +4533,37 @@ free_meta_inode: make_bad_inode(sbi->meta_inode); iput(sbi->meta_inode); sbi->meta_inode = NULL; +free_page_array_cache: + f2fs_destroy_page_array_cache(sbi); +free_xattr_cache: + f2fs_destroy_xattr_caches(sbi); free_io_dummy: mempool_destroy(sbi->write_io_dummy); free_percpu: destroy_percpu_info(sbi); +free_iostat: + f2fs_destroy_iostat(sbi); free_bio_info: for (i = 0; i < NR_PAGE_TYPE; i++) kvfree(sbi->write_io[i]); -#ifdef CONFIG_UNICODE - utf8_unload(sbi->s_encoding); +#if IS_ENABLED(CONFIG_UNICODE) + utf8_unload(sb->s_encoding); + sb->s_encoding = NULL; #endif free_options: #ifdef CONFIG_QUOTA for (i = 0; i < MAXQUOTAS; i++) - kvfree(F2FS_OPTION(sbi).s_qf_names[i]); + kfree(F2FS_OPTION(sbi).s_qf_names[i]); #endif + fscrypt_free_dummy_policy(&F2FS_OPTION(sbi).dummy_enc_policy); kvfree(options); free_sb_buf: - kvfree(raw_super); + kfree(raw_super); free_sbi: if (sbi->s_chksum_driver) crypto_free_shash(sbi->s_chksum_driver); - kvfree(sbi); + kfree(sbi); /* give only one another chance */ if (retry_cnt > 0 && skip_recovery) { @@ -3781,6 +4589,15 @@ static void kill_f2fs_super(struct super_block *sb) f2fs_stop_gc_thread(sbi); f2fs_stop_discard_thread(sbi); +#ifdef CONFIG_F2FS_FS_COMPRESSION + /* + * latter evict_inode() can bypass checking and invalidating + * compress inode cache. + */ + if (test_opt(sbi, COMPRESS_CACHE)) + truncate_inode_pages_final(COMPRESS_MAPPING(sbi)); +#endif + if (is_sbi_flag_set(sbi, SBI_IS_DIRTY) || !is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) { struct cp_control cpc = { @@ -3800,7 +4617,7 @@ static struct file_system_type f2fs_fs_type = { .name = "f2fs", .mount = f2fs_mount, .kill_sb = kill_f2fs_super, - .fs_flags = FS_REQUIRES_DEV, + .fs_flags = FS_REQUIRES_DEV | FS_ALLOW_IDMAP, }; MODULE_ALIAS_FS("f2fs"); @@ -3834,8 +4651,6 @@ static int __init init_f2fs_fs(void) return -EINVAL; } - f2fs_build_trace_ios(); - err = init_inodecache(); if (err) goto fail; @@ -3848,13 +4663,19 @@ static int __init init_f2fs_fs(void) err = f2fs_create_checkpoint_caches(); if (err) goto free_segment_manager_caches; - err = f2fs_create_extent_cache(); + err = f2fs_create_recovery_cache(); if (err) goto free_checkpoint_caches; - err = f2fs_init_sysfs(); + err = f2fs_create_extent_cache(); + if (err) + goto free_recovery_cache; + err = f2fs_create_garbage_collection_cache(); if (err) goto free_extent_cache; - err = register_shrinker(&f2fs_shrinker_info); + err = f2fs_init_sysfs(); + if (err) + goto free_garbage_collection_cache; + err = register_shrinker(&f2fs_shrinker_info, "f2fs-shrinker"); if (err) goto free_sysfs; err = register_filesystem(&f2fs_fs_type); @@ -3864,15 +4685,35 @@ static int __init init_f2fs_fs(void) err = f2fs_init_post_read_processing(); if (err) goto free_root_stats; - err = f2fs_init_bio_entry_cache(); + err = f2fs_init_iostat_processing(); if (err) goto free_post_read; + err = f2fs_init_bio_entry_cache(); + if (err) + goto free_iostat; err = f2fs_init_bioset(); if (err) goto free_bio_enrty_cache; + err = f2fs_init_compress_mempool(); + if (err) + goto free_bioset; + err = f2fs_init_compress_cache(); + if (err) + goto free_compress_mempool; + err = f2fs_create_casefold_cache(); + if (err) + goto free_compress_cache; return 0; +free_compress_cache: + f2fs_destroy_compress_cache(); +free_compress_mempool: + f2fs_destroy_compress_mempool(); +free_bioset: + f2fs_destroy_bioset(); free_bio_enrty_cache: f2fs_destroy_bio_entry_cache(); +free_iostat: + f2fs_destroy_iostat_processing(); free_post_read: f2fs_destroy_post_read_processing(); free_root_stats: @@ -3882,8 +4723,12 @@ free_shrinker: unregister_shrinker(&f2fs_shrinker_info); free_sysfs: f2fs_exit_sysfs(); +free_garbage_collection_cache: + f2fs_destroy_garbage_collection_cache(); free_extent_cache: f2fs_destroy_extent_cache(); +free_recovery_cache: + f2fs_destroy_recovery_cache(); free_checkpoint_caches: f2fs_destroy_checkpoint_caches(); free_segment_manager_caches: @@ -3898,19 +4743,24 @@ fail: static void __exit exit_f2fs_fs(void) { + f2fs_destroy_casefold_cache(); + f2fs_destroy_compress_cache(); + f2fs_destroy_compress_mempool(); f2fs_destroy_bioset(); f2fs_destroy_bio_entry_cache(); + f2fs_destroy_iostat_processing(); f2fs_destroy_post_read_processing(); f2fs_destroy_root_stats(); unregister_filesystem(&f2fs_fs_type); unregister_shrinker(&f2fs_shrinker_info); f2fs_exit_sysfs(); + f2fs_destroy_garbage_collection_cache(); f2fs_destroy_extent_cache(); + f2fs_destroy_recovery_cache(); f2fs_destroy_checkpoint_caches(); f2fs_destroy_segment_manager_caches(); f2fs_destroy_node_manager_caches(); destroy_inodecache(); - f2fs_destroy_trace_ios(); } module_init(init_f2fs_fs) @@ -3919,4 +4769,5 @@ module_exit(exit_f2fs_fs) MODULE_AUTHOR("Samsung Electronics's Praesto Team"); MODULE_DESCRIPTION("Flash Friendly File System"); MODULE_LICENSE("GPL"); +MODULE_SOFTDEP("pre: crc32"); |