Re: [PATCH v2 1/2] xfs: always init fdblocks in mount

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Subject: Re: [PATCH v2 1/2] xfs: always init fdblocks in mount
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Tue, 17 Mar 2020 12:01:22 -0700
Cc: bfoster@xxxxxxxxxx, dchinner@xxxxxxxxxx, sandeen@xxxxxxxxxxx, darrick.wong@xxxxxxxxxx, linux-xfs@xxxxxxxxxxxxxxx, yi.zhang@xxxxxxxxxx, houtao1@xxxxxxxxxx
In-reply-to: <1584428702-127436-2-git-send-email-zhengbin13@huawei.com>
References: <1584428702-127436-1-git-send-email-zhengbin13@huawei.com> <1584428702-127436-2-git-send-email-zhengbin13@huawei.com>

So if we can't trust the sb counters on your fuzzed file system,
how can we trust the AG counters?

Also if we decide that we always want to rebuild and thus always read
in the pagf all the pagf_init checks can go away..

References:
- [PATCH v2 0/2] xfs: always init fdblocks in mount and avoid f_bfree overflow
  - From: Zheng Bin
- [PATCH v2 1/2] xfs: always init fdblocks in mount
  - From: Zheng Bin

Prev by Date: remove the di_version icdinode field v3
Next by Date: [PATCH 1/5] xfs: add a new xfs_sb_version_has_v3inode helper
Previous by thread: [PATCH v2 1/2] xfs: always init fdblocks in mount
Next by thread: [xfs] 3524e9d824: xfstests.xfs.011.fail
Index(es):
- Date
- Thread

[Index of Archives] [XFS Filesystem Development (older mail)] [Linux Filesystem Development] [Linux Audio Users] [Yosemite Trails] [Linux Kernel] [Linux RAID] [Linux SCSI]

Powered by Linux