coreutils
[Top][All Lists]
Advanced

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

Re: improvement for 'ln' to add the '-m' switch


From: Bernhard Voelker
Subject: Re: improvement for 'ln' to add the '-m' switch
Date: Fri, 9 Feb 2024 09:10:49 +0100
User-agent: Mozilla Thunderbird

[+coreutils ML]

On 2/8/24 17:43, aotto wrote:
> Hi, I write to you because of a suggestion from the "https://serverfault.com/"; 
Q&A side

you sent this to my personal email, so you're lucky this didn't land in spam.
Hence I'm hereby forwarding to the coreutils mailing list.

> subtract: add a "-m" switch to "ln" to support "mount -bind srcDir tgtDir"
>
> As reference please read:
>
> → 
https://serverfault.com/questions/1153146/improvement-add-new-switch-to-ln-called-m-to-link-a-directory-as-mount
>
> As original reference please read:
>
> → 
https://stackoverflow.com/questions/77954023/git-check-in-a-file-in-a-symlink-directory-fails/77960139#77960139
>
>
> mfg
>
>    aotto

ln(1) is the userland tool for the link/linkat system calls for hard links
and for the symlink/synlinkat system calls for symbolic aka soft links.
Both hard and soft links are persisted in the file system.

mount(1)-ing is a completely different matter, because it tells the kernel
to well, mount(2), one file system (or part of it) on another place, and
this is gone when the system reboots.

Therefore, I don't see any common point in this, and therefore adding such
a functionality to ln(1) is the wrong thing to do.

Have a nice day,
Berny



reply via email to

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