Received-SPF: pass (google.com: best guess record for domain of linux-ext4-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [HELP] What are the allocated blocks on a newly created ext4 fs ?
To:     Theodore Ts'o <tytso@mit.edu>
Cc:     Artem Blagodarenko <artem.blagodarenko@gmail.com>,
        linux-ext4@vger.kernel.org, adilger.kernel@dilger.ca
References: <b64110a4-8112-a230-2179-5095aa943af9@gmail.com>
 <7FD3148B-1E27-4BFA-965C-9FDC7FC8FD96@gmail.com>
 <22ccd72a-8926-0d12-0fbe-8ad5604d1584@gmail.com>
 <20190610033927.GA15963@mit.edu>
From:   Jianchao Wang <jianchao.wan9@gmail.com>
Message-ID: <683bc83a-a56d-1ecf-aa3d-5aa0aa3ac289@gmail.com>
Date:   Mon, 10 Jun 2019 11:54:35 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:67.0) Gecko/20100101
 Thunderbird/67.0
MIME-Version: 1.0
In-Reply-To: <20190610033927.GA15963@mit.edu>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-ext4-owner@vger.kernel.org
Precedence: bulk

On 2019/6/10 11:39, Theodore Ts'o wrote:
> On Mon, Jun 10, 2019 at 11:25:47AM +0800, Jianchao Wang wrote:
>> Hi Artem 
>>
>> Thanks so much for your help.
>>
>> On 2019/6/6 20:32, Artem Blagodarenko wrote:
>>> Hello Jianchao,
>>>
>>> Not enought input data to give an answer. It depends on mkfs options. For example, if flex_bg option is enabled, then several block groups are tied together as one logical block group; the bitmap spaces and the inode table space in the first block group, so some groups are not totally free just after FS creating.
>>
>> In my environment, there are 16 bgs per flex_bg.
>> The bitmaps and inode table .etc should lay on the first bg of every flex_bg.
>> So I can see there are about 8223 blocks allocated in the 1st bg of every flex_bg.
>>
>> But as you can see in the output of mb_groups, there are some bgs
>> which get allocated about 1024 blocks.
>>
>> I have out figured out what are they for.
> 
> The best way to understand what the blocks are used for is to use the
> dumpe2fs program, e.g:
> 
> % mke2fs -t ext4 -Fq /tmp/foo.img 1T
> % dumpe2fs /tmp/foo.img | more
>     ...
> Group 0: (Blocks 0-32767) csum 0xe6f6 [ITABLE_ZEROED]
>   Primary superblock at 0, Group descriptors at 1-128
>   Reserved GDT blocks at 129-1152
>   Block bitmap at 1153 (+1153), csum 0xb033995a
>   Inode bitmap at 1169 (+1169), csum 0x108d3d73
>   Inode table at 1185-1696 (+1185)
>   23385 free blocks, 8181 free inodes, 2 directories, 8181 unused inodes
>   Free blocks: 9383-32767
>   Free inodes: 12-8192
> Group 1: (Blocks 32768-65535) csum 0xe50b [INODE_UNINIT, BLOCK_UNINIT, ITABLE_ZEROED]
>   Backup superblock at 32768, Group descriptors at 32769-32896
>   Reserved GDT blocks at 32897-33920
>   Block bitmap at 1154 (bg #0 + 1154), csum 0x00000000
>   Inode bitmap at 1170 (bg #0 + 1170), csum 0x00000000
>   Inode table at 1697-2208 (bg #0 + 1697)
>   31615 free blocks, 8192 free inodes, 0 directories, 8192 unused inodes
>   Free blocks: 33921-65535
>   Free inodes: 8193-16384
>     ...
> 

Hi Ted

Many thanks for your help

These allocated blocks should be the reserved GDT blocks.

Many thanks again.
Jianchao