Re: [PATCH v4 09/11] gfs2: iomap direct I/O support

From: Andreas Gruenbacher <agruenba@redhat.com>
To: Christoph Hellwig <hch@lst.de>, Steven Whitehouse <swhiteho@redhat.com>
Cc: cluster-devel <cluster-devel@redhat.com>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>
Subject: Re: [PATCH v4 09/11] gfs2: iomap direct I/O support
Date: Wed, 16 May 2018 22:36:46 +0200	[thread overview]
Message-ID: <CAHc6FU6FBhZQ6qUpz2Eo6nz_BCj_P2NczRnatK2ezdy7xuPXbw@mail.gmail.com> (raw)
In-Reply-To: <20180515073129.GB23384@lst.de>

On 15 May 2018 at 09:31, Christoph Hellwig <hch@lst.de> wrote:
> On Mon, May 14, 2018 at 05:36:22PM +0200, Andreas Gruenbacher wrote:
>> With that, the direct_IO address space operation can be all but
>> eliminated: only a dummy remains which indicates that the filesystem
>> supports direct I/O.
>
> And that dummy can be replaced with the generic noop_direct_IO helper.

Ok, switching to that.

>> +             ret = filemap_write_and_wait_range(mapping, lstart, end);
>> +             if (ret)
>> +                     goto out;
>
> We already do this call in the common code.
>
>> +             if (iov_iter_rw(from) == WRITE)
>> +                     truncate_inode_pages_range(mapping, lstart, end);
>
> Why do you need the truncate_inode_pages_range call here instead of
> invalidate_inode_pages2_range that everyone else uses and that
> the iomap code already does for you?

Indeed, I guess this whole code block can now be discarded. Taking
Steve into the CC since it's him who originally added that.

Thanks,
Andreas