Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Mon, 3 Feb 2020 12:00:51 +0800
From:   chenqiwu <qiwuchen55@gmail.com>
To:     Bernd Schubert <bernd.schubert@fastmail.fm>
Cc:     miklos@szeredi.hu, linux-fsdevel@vger.kernel.org,
        linux-kernel@vger.kernel.org, chenqiwu <chenqiwu@xiaomi.com>
Subject: Re: [PATCH] fuse: Allow parallel DIO reads and check NOWAIT case for
 DIO writes
Message-ID: <20200203040051.GB11846@cqw-OptiPlex-7050>
References: <1580614487-1341-1-git-send-email-qiwuchen55@gmail.com>
 <07d333db-9ed3-2628-673e-cb614c31f29e@fastmail.fm>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <07d333db-9ed3-2628-673e-cb614c31f29e@fastmail.fm>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Sun, Feb 02, 2020 at 10:25:43PM +0100, Bernd Schubert wrote:
> 
> 
> > @@ -1518,6 +1525,9 @@ static ssize_t fuse_direct_read_iter(struct kiocb *iocb, struct iov_iter *to)
> >  
> >  		res = __fuse_direct_read(&io, to, &iocb->ki_pos);
> >  	}
> > +	inode_unlock_shared(inode);
> > +
> > +	file_accessed(iocb->ki_filp);
> 
> 
> Shouldn't the file_accessed() in different patch, with a description? It
> looks totally unrelated to locking?
>
Thanks for your remind! file_accessed() is used to update atime for
every direct read, it's totally unrelated to locking. I will separate
it to another patch.