Re: [PATCH V3 7/7] f2fs: switch to the new mount api

[Chao Yu <chao@kernel.org>]

On 5/14/25 10:33, Hongbo Li wrote:
> 
> 
> On 2025/5/13 16:59, Chao Yu wrote:
>> On 4/24/25 01:08, Eric Sandeen wrote:
>>> From: Hongbo Li <lihongbo22@huawei.com>
>>>
>>> The new mount api will execute .parse_param, .init_fs_context, .get_tree
>>> and will call .remount if remount happened. So we add the necessary
>>> functions for the fs_context_operations. If .init_fs_context is added,
>>> the old .mount should remove.
>>>
>>> See Documentation/filesystems/mount_api.rst for more information.
>>
>> mkfs.f2fs -f -O extra_attr,flexible_inline_xattr /dev/vdb
>> mount -o inline_xattr_size=512 /dev/vdb /mnt/f2fs
>> mount: /mnt/f2fs: wrong fs type, bad option, bad superblock on /dev/vdb, missing codepage or helper program, or other error.
>>         dmesg(1) may have more information after failed mount system call.
>> dmesg
>> [ 1758.202282] F2FS-fs (vdb): Image doesn't support compression
>> [ 1758.202286] F2FS-fs (vdb): inline_xattr_size option should be set with inline_xattr option
>>
>> Eric, Hongbo, can you please take a look at this issue?
>>
> Sorry, we only check the option hold in ctx, we should do the double check in sbi. Or other elegant approaches.
> 
> For the "support compression", is it also the error in this testcase?

I think so, I checked this w/ additional logs, and found this:

#define F2FS_MOUNT_INLINE_XATTR_SIZE	0x00080000
#define F2FS_SPEC_compress_chksum		(1 << 19) /* equal to 0x00080000)

	if (!f2fs_sb_has_compression(sbi)) {
		if (test_compression_spec(ctx->opt_mask) ||
			ctx_test_opt(ctx, F2FS_MOUNT_COMPRESS_CACHE))
			f2fs_info(sbi, "Image doesn't support compression");
		clear_compression_spec(ctx);
		ctx->opt_mask &= ~F2FS_MOUNT_COMPRESS_CACHE;
		return 0;
	}

We should use test_compression_spec(ctx->spec_mask) instead of
test_compression_spec(ctx->opt_mask) to check special mask of compression
option?

Thanks,

> 
> Thanks,
> Hongbo
> 
>> Thanks,
>>
>>>
>>> Signed-off-by: Hongbo Li <lihongbo22@huawei.com>
>>> [sandeen: forward port]
>>> Signed-off-by: Eric Sandeen <sandeen@redhat.com>
>>> ---
>>>   fs/f2fs/super.c | 156 +++++++++++++++++++-----------------------------
>>>   1 file changed, 62 insertions(+), 94 deletions(-)
>>>
>>> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
>>> index 37497fd80bb9..041bd6c482a0 100644
>>> --- a/fs/f2fs/super.c
>>> +++ b/fs/f2fs/super.c
>>> @@ -1141,47 +1141,6 @@ static int f2fs_parse_param(struct fs_context *fc, struct fs_parameter *param)
>>>       return 0;
>>>   }
>>>   -static int parse_options(struct fs_context *fc, char *options)
>>> -{
>>> -    struct fs_parameter param;
>>> -    char *key;
>>> -    int ret;
>>> -
>>> -    if (!options)
>>> -        return 0;
>>> -
>>> -    while ((key = strsep(&options, ",")) != NULL) {
>>> -        if (*key) {
>>> -            size_t v_len = 0;
>>> -            char *value = strchr(key, '=');
>>> -
>>> -            param.type = fs_value_is_flag;
>>> -            param.string = NULL;
>>> -
>>> -            if (value) {
>>> -                if (value == key)
>>> -                    continue;
>>> -
>>> -                *value++ = 0;
>>> -                v_len = strlen(value);
>>> -                param.string = kmemdup_nul(value, v_len, GFP_KERNEL);
>>> -                if (!param.string)
>>> -                    return -ENOMEM;
>>> -                param.type = fs_value_is_string;
>>> -            }
>>> -
>>> -            param.key = key;
>>> -            param.size = v_len;
>>> -
>>> -            ret = f2fs_parse_param(fc, &param);
>>> -            kfree(param.string);
>>> -            if (ret < 0)
>>> -                return ret;
>>> -        }
>>> -    }
>>> -    return 0;
>>> -}
>>> -
>>>   /*
>>>    * Check quota settings consistency.
>>>    */
>>> @@ -2583,13 +2542,12 @@ static void f2fs_enable_checkpoint(struct f2fs_sb_info *sbi)
>>>       f2fs_flush_ckpt_thread(sbi);
>>>   }
>>>   -static int f2fs_remount(struct super_block *sb, int *flags, char *data)
>>> +static int __f2fs_remount(struct fs_context *fc, struct super_block *sb)
>>>   {
>>>       struct f2fs_sb_info *sbi = F2FS_SB(sb);
>>>       struct f2fs_mount_info org_mount_opt;
>>> -    struct f2fs_fs_context ctx;
>>> -    struct fs_context fc;
>>>       unsigned long old_sb_flags;
>>> +    unsigned int flags = fc->sb_flags;
>>>       int err;
>>>       bool need_restart_gc = false, need_stop_gc = false;
>>>       bool need_restart_flush = false, need_stop_flush = false;
>>> @@ -2635,7 +2593,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
>>>   #endif
>>>         /* recover superblocks we couldn't write due to previous RO mount */
>>> -    if (!(*flags & SB_RDONLY) && is_sbi_flag_set(sbi, SBI_NEED_SB_WRITE)) {
>>> +    if (!(flags & SB_RDONLY) && is_sbi_flag_set(sbi, SBI_NEED_SB_WRITE)) {
>>>           err = f2fs_commit_super(sbi, false);
>>>           f2fs_info(sbi, "Try to recover all the superblocks, ret: %d",
>>>                 err);
>>> @@ -2645,21 +2603,11 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
>>>         default_options(sbi, true);
>>>   -    memset(&fc, 0, sizeof(fc));
>>> -    memset(&ctx, 0, sizeof(ctx));
>>> -    fc.fs_private = &ctx;
>>> -    fc.purpose = FS_CONTEXT_FOR_RECONFIGURE;
>>> -
>>> -    /* parse mount options */
>>> -    err = parse_options(&fc, data);
>>> -    if (err)
>>> -        goto restore_opts;
>>> -
>>> -    err = f2fs_check_opt_consistency(&fc, sb);
>>> +    err = f2fs_check_opt_consistency(fc, sb);
>>>       if (err < 0)
>>>           goto restore_opts;
>>>   -    f2fs_apply_options(&fc, sb);
>>> +    f2fs_apply_options(fc, sb);
>>>     #ifdef CONFIG_BLK_DEV_ZONED
>>>       if (f2fs_sb_has_blkzoned(sbi) &&
>>> @@ -2678,20 +2626,20 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
>>>        * Previous and new state of filesystem is RO,
>>>        * so skip checking GC and FLUSH_MERGE conditions.
>>>        */
>>> -    if (f2fs_readonly(sb) && (*flags & SB_RDONLY))
>>> +    if (f2fs_readonly(sb) && (flags & SB_RDONLY))
>>>           goto skip;
>>>   -    if (f2fs_dev_is_readonly(sbi) && !(*flags & SB_RDONLY)) {
>>> +    if (f2fs_dev_is_readonly(sbi) && !(flags & SB_RDONLY)) {
>>>           err = -EROFS;
>>>           goto restore_opts;
>>>       }
>>>     #ifdef CONFIG_QUOTA
>>> -    if (!f2fs_readonly(sb) && (*flags & SB_RDONLY)) {
>>> +    if (!f2fs_readonly(sb) && (flags & SB_RDONLY)) {
>>>           err = dquot_suspend(sb, -1);
>>>           if (err < 0)
>>>               goto restore_opts;
>>> -    } else if (f2fs_readonly(sb) && !(*flags & SB_RDONLY)) {
>>> +    } else if (f2fs_readonly(sb) && !(flags & SB_RDONLY)) {
>>>           /* dquot_resume needs RW */
>>>           sb->s_flags &= ~SB_RDONLY;
>>>           if (sb_any_quota_suspended(sb)) {
>>> @@ -2747,7 +2695,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
>>>           goto restore_opts;
>>>       }
>>>   -    if ((*flags & SB_RDONLY) && test_opt(sbi, DISABLE_CHECKPOINT)) {
>>> +    if ((flags & SB_RDONLY) && test_opt(sbi, DISABLE_CHECKPOINT)) {
>>>           err = -EINVAL;
>>>           f2fs_warn(sbi, "disabling checkpoint not compatible with read-only");
>>>           goto restore_opts;
>>> @@ -2758,7 +2706,7 @@ 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) ||
>>> +    if ((flags & SB_RDONLY) ||
>>>               (F2FS_OPTION(sbi).bggc_mode == BGGC_MODE_OFF &&
>>>               !test_opt(sbi, GC_MERGE))) {
>>>           if (sbi->gc_thread) {
>>> @@ -2772,7 +2720,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
>>>           need_stop_gc = true;
>>>       }
>>>   -    if (*flags & SB_RDONLY) {
>>> +    if (flags & SB_RDONLY) {
>>>           sync_inodes_sb(sb);
>>>             set_sbi_flag(sbi, SBI_IS_DIRTY);
>>> @@ -2785,7 +2733,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
>>>        * We stop issue flush thread if FS is mounted as RO
>>>        * or if flush_merge is not passed in mount option.
>>>        */
>>> -    if ((*flags & SB_RDONLY) || !test_opt(sbi, FLUSH_MERGE)) {
>>> +    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;
>>> @@ -2827,7 +2775,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
>>>        * triggered while remount and we need to take care of it before
>>>        * returning from remount.
>>>        */
>>> -    if ((*flags & SB_RDONLY) || test_opt(sbi, DISABLE_CHECKPOINT) ||
>>> +    if ((flags & SB_RDONLY) || test_opt(sbi, DISABLE_CHECKPOINT) ||
>>>               !test_opt(sbi, MERGE_CHECKPOINT)) {
>>>           f2fs_stop_ckpt_thread(sbi);
>>>       } else {
>>> @@ -2854,7 +2802,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
>>>           (test_opt(sbi, POSIX_ACL) ? SB_POSIXACL : 0);
>>>         limit_reserve_root(sbi);
>>> -    *flags = (*flags & ~SB_LAZYTIME) | (sb->s_flags & SB_LAZYTIME);
>>> +    fc->sb_flags = (flags & ~SB_LAZYTIME) | (sb->s_flags & SB_LAZYTIME);
>>>         sbi->umount_lock_holder = NULL;
>>>       return 0;
>>> @@ -3523,7 +3471,6 @@ static const struct super_operations f2fs_sops = {
>>>       .freeze_fs    = f2fs_freeze,
>>>       .unfreeze_fs    = f2fs_unfreeze,
>>>       .statfs        = f2fs_statfs,
>>> -    .remount_fs    = f2fs_remount,
>>>       .shutdown    = f2fs_shutdown,
>>>   };
>>>   @@ -4745,16 +4692,13 @@ static void f2fs_tuning_parameters(struct f2fs_sb_info *sbi)
>>>       sbi->readdir_ra = true;
>>>   }
>>>   -static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
>>> +static int f2fs_fill_super(struct super_block *sb, struct fs_context *fc)
>>>   {
>>>       struct f2fs_sb_info *sbi;
>>>       struct f2fs_super_block *raw_super;
>>> -    struct f2fs_fs_context ctx;
>>> -    struct fs_context fc;
>>>       struct inode *root;
>>>       int err;
>>>       bool skip_recovery = false, need_fsck = false;
>>> -    char *options = NULL;
>>>       int recovery, i, valid_super_block;
>>>       struct curseg_info *seg_i;
>>>       int retry_cnt = 1;
>>> @@ -4767,9 +4711,6 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
>>>       raw_super = NULL;
>>>       valid_super_block = -1;
>>>       recovery = 0;
>>> -    memset(&fc, 0, sizeof(fc));
>>> -    memset(&ctx, 0, sizeof(ctx));
>>> -    fc.fs_private = &ctx;
>>>         /* allocate memory for f2fs-specific super block info */
>>>       sbi = kzalloc(sizeof(struct f2fs_sb_info), GFP_KERNEL);
>>> @@ -4820,22 +4761,12 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
>>>                           sizeof(raw_super->uuid));
>>>         default_options(sbi, false);
>>> -    /* parse mount options */
>>> -    options = kstrdup((const char *)data, GFP_KERNEL);
>>> -    if (data && !options) {
>>> -        err = -ENOMEM;
>>> -        goto free_sb_buf;
>>> -    }
>>> -
>>> -    err = parse_options(&fc, options);
>>> -    if (err)
>>> -        goto free_options;
>>>   -    err = f2fs_check_opt_consistency(&fc, sb);
>>> +    err = f2fs_check_opt_consistency(fc, sb);
>>>       if (err < 0)
>>> -        goto free_options;
>>> +        goto free_sb_buf;
>>>   -    f2fs_apply_options(&fc, sb);
>>> +    f2fs_apply_options(fc, sb);
>>>         sb->s_maxbytes = max_file_blocks(NULL) <<
>>>                   le32_to_cpu(raw_super->log_blocksize);
>>> @@ -5160,7 +5091,6 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
>>>           if (err)
>>>               goto sync_free_meta;
>>>       }
>>> -    kvfree(options);
>>>         /* recover broken superblock */
>>>       if (recovery) {
>>> @@ -5255,7 +5185,6 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
>>>           kfree(F2FS_OPTION(sbi).s_qf_names[i]);
>>>   #endif
>>>       fscrypt_free_dummy_policy(&F2FS_OPTION(sbi).dummy_enc_policy);
>>> -    kvfree(options);
>>>   free_sb_buf:
>>>       kfree(raw_super);
>>>   free_sbi:
>>> @@ -5271,14 +5200,39 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
>>>       return err;
>>>   }
>>>   -static struct dentry *f2fs_mount(struct file_system_type *fs_type, int flags,
>>> -            const char *dev_name, void *data)
>>> +static int f2fs_get_tree(struct fs_context *fc)
>>>   {
>>> -    return mount_bdev(fs_type, flags, dev_name, data, f2fs_fill_super);
>>> +    return get_tree_bdev(fc, f2fs_fill_super);
>>> +}
>>> +
>>> +static int f2fs_reconfigure(struct fs_context *fc)
>>> +{
>>> +    struct super_block *sb = fc->root->d_sb;
>>> +
>>> +    return __f2fs_remount(fc, sb);
>>> +}
>>> +
>>> +static void f2fs_fc_free(struct fs_context *fc)
>>> +{
>>> +    struct f2fs_fs_context *ctx = fc->fs_private;
>>> +    int i;
>>> +
>>> +    if (!ctx)
>>> +        return;
>>> +
>>> +#ifdef CONFIG_QUOTA
>>> +    for (i = 0; i < MAXQUOTAS; i++)
>>> +        kfree(F2FS_CTX_INFO(ctx).s_qf_names[i]);
>>> +#endif
>>> +    fscrypt_free_dummy_policy(&F2FS_CTX_INFO(ctx).dummy_enc_policy);
>>> +    kfree(ctx);
>>>   }
>>>     static const struct fs_context_operations f2fs_context_ops = {
>>>       .parse_param    = f2fs_parse_param,
>>> +    .get_tree    = f2fs_get_tree,
>>> +    .reconfigure = f2fs_reconfigure,
>>> +    .free    = f2fs_fc_free,
>>>   };
>>>     static void kill_f2fs_super(struct super_block *sb)
>>> @@ -5322,10 +5276,24 @@ static void kill_f2fs_super(struct super_block *sb)
>>>       }
>>>   }
>>>   +static int f2fs_init_fs_context(struct fs_context *fc)
>>> +{
>>> +    struct f2fs_fs_context *ctx;
>>> +
>>> +    ctx = kzalloc(sizeof(struct f2fs_fs_context), GFP_KERNEL);
>>> +    if (!ctx)
>>> +        return -ENOMEM;
>>> +
>>> +    fc->fs_private = ctx;
>>> +    fc->ops = &f2fs_context_ops;
>>> +
>>> +    return 0;
>>> +}
>>> +
>>>   static struct file_system_type f2fs_fs_type = {
>>>       .owner        = THIS_MODULE,
>>>       .name        = "f2fs",
>>> -    .mount        = f2fs_mount,
>>> +    .init_fs_context = f2fs_init_fs_context,
>>>       .kill_sb    = kill_f2fs_super,
>>>       .fs_flags    = FS_REQUIRES_DEV | FS_ALLOW_IDMAP,
>>>   };
>>