coreutils
[Top][All Lists]
Advanced

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

Re: df command reports incorrect usage


From: Rob Landley
Subject: Re: df command reports incorrect usage
Date: Thu, 24 Mar 2022 11:12:40 -0500
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.14.0

On 3/24/22 10:26 AM, Fariya F wrote:
> Thank you for all the inputs. 
> 
> I have a question what is the meaning of overhead block (its different from
> reserved block) and how the number of overhead blocks is  calculated?

Not really a df/coreutils question. The command fetches info from the stat() and
statvfs() system calls (man 2 stat, man 2 statvfs) and neither structure has an
"overhead" field. I assume you're referring to:

  https://github.com/torvalds/linux/blob/master/fs/ext2/super.c#L1406

Which would be a kernel issue. Try asking on lkml or one of the more
domain-specific kernel lists like:

  http://vger.kernel.org/vger-lists.html#linux-fsdevel

Or more likely:

  http://vger.kernel.org/vger-lists.html#linux-ext4

(Since ext4 is theoretically backwards compatible with ext3 and ext2 they
collapsed the dev community together and that's the surviving list. The ext2
driver survives for embedded users. The ext3 driver got deleted once ext4 was
stable.)

> The
> problem is the num of overhead blocks are more than the total blocks for that
> partition and that's why the total blocks - overhead blocks returns a huge
> value. (Negative number). 

Yeah, that'd do it.

> Anyone could help as to how overhead blocks are Calculated? What could go 
> wrong
> in this calculation? 

Above link goes to the kernel code that's doing it. I'm not familiar with the
driver logic in question.

Rob



reply via email to

[Prev in Thread] Current Thread [Next in Thread]