Date:   Sun, 9 Jan 2022 23:53:20 -0800
From:   Christoph Hellwig <hch@infradead.org>
To:     Matthew Wilcox <willy@infradead.org>
Cc:     Amir Goldstein <amir73il@gmail.com>,
        David Howells <dhowells@redhat.com>, linux-cachefs@redhat.com,
        Trond Myklebust <trondmy@hammerspace.com>,
        Anna Schumaker <anna.schumaker@netapp.com>,
        Steve French <sfrench@samba.org>,
        Dominique Martinet <asmadeus@codewreck.org>,
        Jeff Layton <jlayton@kernel.org>,
        Alexander Viro <viro@zeniv.linux.org.uk>,
        Omar Sandoval <osandov@osandov.com>,
        JeffleXu <jefflexu@linux.alibaba.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        linux-afs@lists.infradead.org,
        Linux NFS Mailing List <linux-nfs@vger.kernel.org>,
        CIFS <linux-cifs@vger.kernel.org>,
        ceph-devel <ceph-devel@vger.kernel.org>,
        v9fs-developer@lists.sourceforge.net,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v4 38/68] vfs, cachefiles: Mark a backing file in use
 with an inode flag
Message-ID: <Ydvl8Dk8z0mF0KFl@infradead.org>
References: <164021479106.640689.17404516570194656552.stgit@warthog.procyon.org.uk>
 <164021541207.640689.564689725898537127.stgit@warthog.procyon.org.uk>
 <CAOQ4uxjEcvffv=rNXS-r+NLz+=6yk4abRuX_AMq9v-M4nf_PtA@mail.gmail.com>
 <Ydk6jWmFH6TZLPZq@casper.infradead.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Ydk6jWmFH6TZLPZq@casper.infradead.org>
Precedence: bulk

On Sat, Jan 08, 2022 at 07:17:33AM +0000, Matthew Wilcox wrote:
> On Sat, Jan 08, 2022 at 09:08:57AM +0200, Amir Goldstein wrote:
> > > +#define S_KERNEL_FILE  (1 << 17) /* File is in use by the kernel (eg. fs/cachefiles) */
> > 
> > Trying to brand this flag as a generic "in use by kernel" is misleading.
> > Modules other than cachefiles cannot set/clear this flag, because then
> > cachefiles won't know that it is allowed to set/clear the flag.
> 
> Huh?  If some other kernel module sets it, cachefiles will try to set it,
> see it's already set, and fail.  Presumably cachefiles does not go round
> randomly "unusing" files that it has not previously started using.
> 
> I mean, yes, obviously, it's a kernel module, it can set and clear
> whatever flags it likes on any inode in the system, but conceptually,
> it's an advisory whole-file lock.

So let's name it that way.  We have plenty of files in kernel use using
filp_open and this flag very obviously means something else.