From: "wangxiaojun (N)" <wangxiaojun11@huawei.com>
To: Chao Yu <chao@kernel.org>, <jaegeuk@kernel.org>
Cc: linux-f2fs-devel@lists.sourceforge.net
Subject: Re: [f2fs-dev] [PATCH] f2fs-tools: rebuild the quota inode if it is corrupted
Date: Mon, 19 Jul 2021 09:43:40 +0800 [thread overview]
Message-ID: <ed0a9452-b29f-6aa8-c52b-bc584fbaec4a@huawei.com> (raw)
In-Reply-To: <776a5989-3309-3a2e-40fb-6c12fef5d9dc@kernel.org>
在 2021/7/17 16:16, Chao Yu 写道:
> On 2021/7/14 10:00, Wang Xiaojun wrote:
>> If the following process returns an error,
>> the quota inode, not the quota file, is damaged.
>> (fsck_chk_quota_node-->fsck_chk_node_blk-->sanity_check_nid)
>> The fsck does not have a process to rebuild the quota inode.
>>
>> Because sanity_check_nid is not passed, fsck->nat_area_bitmap
>> can not be cleared, and then the NAT of quota will be nullify
>> during fix_nat_entries.
>>
>> During the next fsck check, the quota inode check fails
>> because the address of the quota inode changes to 0.
>> In addition, in fsck_chk_quota_files-->f2fs_filesize_update,
>> data is written to address 0.
>>
>> Therefore, when the quota inode is corrupted, we need to rebuild it.
>>
>> Signed-off-by: Wang Xiaojun <wangxiaojun11@huawei.com>
>> ---
>> fsck/fsck.c | 2 ++
>> fsck/fsck.h | 1 +
>> fsck/node.c | 38 ++++++++++++++++++++++++++++++++++++++
>> include/f2fs_fs.h | 40 ++++++++++++++++++++++++++++++++++++++++
>> mkfs/f2fs_format.c | 31 +------------------------------
>> 5 files changed, 82 insertions(+), 30 deletions(-)
>>
>> diff --git a/fsck/fsck.c b/fsck/fsck.c
>> index 6ca85f0..af6d332 100644
>> --- a/fsck/fsck.c
>> +++ b/fsck/fsck.c
>> @@ -1899,6 +1899,8 @@ int fsck_chk_quota_node(struct f2fs_sb_info *sbi)
>> ASSERT_MSG("wrong quota inode, qtype [%d] ino [0x%x]",
>> qtype, ino);
>> qf_szchk_type[qtype] = QF_SZCHK_ERR;
>> + if (c.fix_on)
>> + f2fs_rebuild_qf_inode(sbi, qtype);
>> }
>> }
>> cur_qtype = -1;
>> diff --git a/fsck/fsck.h b/fsck/fsck.h
>> index d79afef..67390f0 100644
>> --- a/fsck/fsck.h
>> +++ b/fsck/fsck.h
>> @@ -286,6 +286,7 @@ void f2fs_alloc_nid(struct f2fs_sb_info *, nid_t *);
>> void set_data_blkaddr(struct dnode_of_data *);
>> block_t new_node_block(struct f2fs_sb_info *,
>> struct dnode_of_data *, unsigned int);
>> +int f2fs_rebuild_qf_inode(struct f2fs_sb_info *sbi, int qtype);
>> /* segment.c */
>> struct quota_file;
>> diff --git a/fsck/node.c b/fsck/node.c
>> index c7988cb..d8d01db 100644
>> --- a/fsck/node.c
>> +++ b/fsck/node.c
>> @@ -40,6 +40,44 @@ void f2fs_release_nid(struct f2fs_sb_info *sbi,
>> nid_t nid)
>> f2fs_clear_bit(nid, nm_i->nid_bitmap);
>> }
>> +int f2fs_rebuild_qf_inode(struct f2fs_sb_info *sbi, int qtype)
>> +{
>> + struct f2fs_node *raw_node = NULL;
>> + struct f2fs_super_block *sb = F2FS_RAW_SUPER(sbi);
>> + struct f2fs_summary sum;
>> + struct node_info ni;
>> + nid_t ino = QUOTA_INO(sb, qtype);
>> + block_t blkaddr = NULL_ADDR;
>> + int ret = 0;
>> +
>> + raw_node = calloc(F2FS_BLKSIZE, 1);
>> + if (raw_node == NULL) {
>> + MSG(1, "\tError: Calloc Failed for raw_node!!!\n");
>> + return -ENOMEM;
>> + }
>> + f2fs_init_qf_inode(sb, raw_node, qtype);
>> + raw_node->footer.cp_ver = F2FS_CKPT(sbi)->checkpoint_ver;
>
> footer.cp_ver not simply equal to .checkpoint_ver, it depends on
> CP_CRC_RECOVERY_FLAG flag.
>
Got it.
Thanks,
>> +
>> + get_node_info(sbi, ino, &ni);
>> + set_summary(&sum, ino, 0, ni.version);
>> + ret = reserve_new_block(sbi, &blkaddr, &sum, CURSEG_HOT_NODE, 1);
>> + if (ret)
>> + goto err_out;
>> +
>> + ret = write_inode(raw_node, blkaddr);
>> + if (ret < 0) {
>> + MSG(1, "\tError: While rebuilding the quota inode to disk!\n");
>> + goto err_out;
>> + }
>> + update_nat_blkaddr(sbi, ino, ino, blkaddr);
>> +
>> + f2fs_clear_bit(ino, F2FS_FSCK(sbi)->nat_area_bitmap);
>> + f2fs_set_bit(ino, NM_I(sbi)->nid_bitmap);
>> +err_out:
>> + free(raw_node);
>> + return ret;
>
> Better to print debug message like we did in f2fs_write_qf_inode()?
Got it.
Thanks,
>
>> +}
>> +
>> void set_data_blkaddr(struct dnode_of_data *dn)
>> {
>> __le32 *addr_array;
>> diff --git a/include/f2fs_fs.h b/include/f2fs_fs.h
>> index 45f7257..87bbd3c 100644
>> --- a/include/f2fs_fs.h
>> +++ b/include/f2fs_fs.h
>> @@ -20,6 +20,7 @@
>> #include <stdio.h>
>> #include <stdlib.h>
>> #include <string.h>
>> +#include <time.h>
>> #ifdef HAVE_CONFIG_H
>> #include <config.h>
>> #endif
>> @@ -1554,6 +1555,45 @@ static inline void show_version(const char *prog)
>> #endif
>> }
>> +static inline void f2fs_init_qf_inode(struct f2fs_super_block *sb,
>> + struct f2fs_node *raw_node, int qtype)
>> +{
>> + raw_node->footer.nid = sb->qf_ino[qtype];
>> + raw_node->footer.ino = sb->qf_ino[qtype];
>> + raw_node->footer.cp_ver = cpu_to_le64(1);
>> + raw_node->i.i_mode = cpu_to_le16(0x8180);
>> + raw_node->i.i_links = cpu_to_le32(1);
>> + raw_node->i.i_uid = cpu_to_le32(c.root_uid);
>> + raw_node->i.i_gid = cpu_to_le32(c.root_gid);
>> +
>> + raw_node->i.i_size = cpu_to_le64(1024 * 6); /* Hard coded */
>> + raw_node->i.i_blocks = cpu_to_le64(1);
>> +
>> + raw_node->i.i_atime = cpu_to_le32(time(NULL));
>> + raw_node->i.i_atime_nsec = 0;
>> + raw_node->i.i_ctime = cpu_to_le32(time(NULL));
>> + raw_node->i.i_ctime_nsec = 0;
>> + raw_node->i.i_mtime = cpu_to_le32(time(NULL));
>> + raw_node->i.i_mtime_nsec = 0;
>> + raw_node->i.i_generation = 0;
>> + raw_node->i.i_xattr_nid = 0;
>> + raw_node->i.i_flags = FS_IMMUTABLE_FL;
>> + raw_node->i.i_current_depth = cpu_to_le32(0);
>> + raw_node->i.i_dir_level = DEF_DIR_LEVEL;
>> +
>> + if (c.feature & cpu_to_le32(F2FS_FEATURE_EXTRA_ATTR)) {
>> + raw_node->i.i_inline = F2FS_EXTRA_ATTR;
>> + raw_node->i.i_extra_isize = cpu_to_le16(calc_extra_isize());
>> + }
>> +
>> + if (c.feature & cpu_to_le32(F2FS_FEATURE_PRJQUOTA))
>> + raw_node->i.i_projid = cpu_to_le32(F2FS_DEF_PROJID);
>> +
>> + raw_node->i.i_ext.fofs = 0;
>> + raw_node->i.i_ext.blk_addr = 0;
>> + raw_node->i.i_ext.len = 0;
>> +}
>> +
>> struct feature {
>> char *name;
>> u32 mask;
>> diff --git a/mkfs/f2fs_format.c b/mkfs/f2fs_format.c
>> index 2132852..f8d0bce 100644
>> --- a/mkfs/f2fs_format.c
>> +++ b/mkfs/f2fs_format.c
>> @@ -1373,42 +1373,16 @@ static int f2fs_write_qf_inode(int qtype)
>> MSG(1, "\tError: Calloc Failed for raw_node!!!\n");
>> return -1;
>> }
>> + f2fs_init_qf_inode(sb, raw_node, qtype);
>> - raw_node->footer.nid = sb->qf_ino[qtype];
>> - raw_node->footer.ino = sb->qf_ino[qtype];
>> - raw_node->footer.cp_ver = cpu_to_le64(1);
>> raw_node->footer.next_blkaddr = cpu_to_le32(
>> get_sb(main_blkaddr) +
>> c.cur_seg[CURSEG_HOT_NODE] *
>> c.blks_per_seg + 1 + qtype + 1);
>> -
>> - raw_node->i.i_mode = cpu_to_le16(0x8180);
>> - raw_node->i.i_links = cpu_to_le32(1);
>> - raw_node->i.i_uid = cpu_to_le32(c.root_uid);
>> - raw_node->i.i_gid = cpu_to_le32(c.root_gid);
>> -
>> - raw_node->i.i_size = cpu_to_le64(1024 * 6); /* Hard coded */
>> raw_node->i.i_blocks = cpu_to_le64(1 + QUOTA_DATA(qtype));
>> -
>> raw_node->i.i_atime = cpu_to_le32(mkfs_time);
>
> mkfs_time or time(NULL) can be a parameter of f2fs_init_qf_inode()?
>
> Thanks,
Makes sense.
I will send a v2 patch.
Thanks,
>
>> - raw_node->i.i_atime_nsec = 0;
>> raw_node->i.i_ctime = cpu_to_le32(mkfs_time);
>> - raw_node->i.i_ctime_nsec = 0;
>> raw_node->i.i_mtime = cpu_to_le32(mkfs_time);
>> - raw_node->i.i_mtime_nsec = 0;
>> - raw_node->i.i_generation = 0;
>> - raw_node->i.i_xattr_nid = 0;
>> - raw_node->i.i_flags = FS_IMMUTABLE_FL;
>> - raw_node->i.i_current_depth = cpu_to_le32(0);
>> - raw_node->i.i_dir_level = DEF_DIR_LEVEL;
>> -
>> - if (c.feature & cpu_to_le32(F2FS_FEATURE_EXTRA_ATTR)) {
>> - raw_node->i.i_inline = F2FS_EXTRA_ATTR;
>> - raw_node->i.i_extra_isize = cpu_to_le16(calc_extra_isize());
>> - }
>> -
>> - if (c.feature & cpu_to_le32(F2FS_FEATURE_PRJQUOTA))
>> - raw_node->i.i_projid = cpu_to_le32(F2FS_DEF_PROJID);
>> data_blk_nor = get_sb(main_blkaddr) +
>> c.cur_seg[CURSEG_HOT_DATA] * c.blks_per_seg + 1;
>> @@ -1434,9 +1408,6 @@ static int f2fs_write_qf_inode(int qtype)
>> for (i = 0; i < QUOTA_DATA(qtype); i++)
>> raw_node->i.i_addr[get_extra_isize(raw_node) + i] =
>> cpu_to_le32(data_blk_nor + i);
>> - raw_node->i.i_ext.fofs = 0;
>> - raw_node->i.i_ext.blk_addr = 0;
>> - raw_node->i.i_ext.len = 0;
>> main_area_node_seg_blk_offset = get_sb(main_blkaddr);
>> main_area_node_seg_blk_offset += c.cur_seg[CURSEG_HOT_NODE] *
>>
> .
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
prev parent reply other threads:[~2021-07-19 1:43 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-14 2:00 [f2fs-dev] [PATCH] f2fs-tools: rebuild the quota inode if it is corrupted Wang Xiaojun
2021-07-17 8:16 ` Chao Yu
2021-07-19 1:43 ` wangxiaojun (N) [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ed0a9452-b29f-6aa8-c52b-bc584fbaec4a@huawei.com \
--to=wangxiaojun11@huawei.com \
--cc=chao@kernel.org \
--cc=jaegeuk@kernel.org \
--cc=linux-f2fs-devel@lists.sourceforge.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).