[Top] [All Lists]

Re: [PATCH 9/10] xfs: implement bulkstat per allocation group

To: Mark Tinguely <tinguely@xxxxxxx>
Subject: Re: [PATCH 9/10] xfs: implement bulkstat per allocation group
From: Jeff Liu <jeff.liu@xxxxxxxxxx>
Date: Tue, 31 Dec 2013 18:00:08 +0800
Cc: "xfs@xxxxxxxxxxx" <xfs@xxxxxxxxxxx>
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <52C1B900.8080002@xxxxxxx>
References: <52BEB47B.3090001@xxxxxxxxxx> <52C1B900.8080002@xxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.1.0
On 2013å12æ31æ 02:18, Mark Tinguely wrote:
> On 12/28/13 05:22, Jeff Liu wrote:
>> From: Jie Liu<jeff.liu@xxxxxxxxxx>
>> With xfs_perag_bulkstat(), we can return stat information in bulk by
>> inode as per the given allocation group number and start inode number
>> in it via a new ioctl(2) in the future.
>> Refactor xfs_bulkstat() with it.
>> Signed-off-by: Jie Liu<jeff.liu@xxxxxxxxxx>
>> ---
> causes xfs/183 to assert for me:
> XFS: Assertion failed: fs_is_ok, file: /root/xfs/fs/xfs/xfs_itable.c,
> line: 409
Oops!! I guess the assertion is failed at:

+                       error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
+                       if (error)
+                               break;
+                       XFS_WANT_CORRUPTED_GOTO(i == 1, error0);

Which means I have to take care of those non-primary AGs on an new initialized
empty-filesystem as there has no inode chunk has been allocated...


<Prev in Thread] Current Thread [Next in Thread>