Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.36 as permitted sender) client-ip=23.128.96.36;
Date:   Fri, 24 Nov 2023 08:26:21 +0000
From:   Al Viro <viro@zeniv.linux.org.uk>
To:     Amir Goldstein <amir73il@gmail.com>
Cc:     linux-fsdevel@vger.kernel.org,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Christian Brauner <brauner@kernel.org>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH 09/20] [software coproarchaeology] dentry.h: kill a
 mysterious comment
Message-ID: <20231124082621.GW38156@ZenIV>
References: <20231124060553.GA575483@ZenIV>
 <20231124060644.576611-1-viro@zeniv.linux.org.uk>
 <20231124060644.576611-9-viro@zeniv.linux.org.uk>
 <CAOQ4uxgRiQCG_Q5TP+05_N4V=iFTemzGTd62ePgAgotK52EAAQ@mail.gmail.com>
 <20231124081141.GV38156@ZenIV>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20231124081141.GV38156@ZenIV>
Sender: Al Viro <viro@ftp.linux.org.uk>
Precedence: bulk

On Fri, Nov 24, 2023 at 08:11:41AM +0000, Al Viro wrote:
> On Fri, Nov 24, 2023 at 09:37:16AM +0200, Amir Goldstein wrote:
> 
> > Since you like pre-git archeology...
> > 
> > Mind digging up what this comment in fs.h is about:
> > 
> > /* needed for stackable file system support */
> > extern loff_t default_llseek(struct file *file, loff_t offset, int whence);
> 
> Umm...  I think it was about ecryptfs and its ilk, but it was a long
> time ago...
> 
> <looks>
> 
> 2.3.99pre6, along with exporting it:
> -/* for stackable file systems (lofs, wrapfs, etc.) */
> -EXPORT_SYMBOL(add_to_page_cache);
> +/* for stackable file systems (lofs, wrapfs, cryptfs, etc.) */
> +EXPORT_SYMBOL(default_llseek);
> +EXPORT_SYMBOL(dentry_open);
> 
> Back then ->llseek == NULL used to mean default_llseek; that had
> been changed much later.  And that was before vfs_llseek() as well,
> so any layered filesystem had to open-code it - which required
> default_llseek().
> 
> The comment is certainly stale, though - stackable filesystems do *not*
> need it (vfs_llseek() is there), but every file_operation that used
> to have NULL ->llseek does.
> 
> > Or we can just remove it without digging up what the comment used
> > to refer to ;)
> 
> Too late - it will have to be removed with that ;-)

BTW, there's this, covering more than just BK times:

git://git.kernel.org/pub/scm/linux/kernel/git/history/history.git/