Re: [PATCH v8 06/10] iomap: Add page_write_end iomap hook

From: "Andreas Grünbacher" <andreas.gruenbacher@gmail.com>
To: dsterba@suse.cz
Cc: Andreas Gruenbacher <agruenba@redhat.com>,
	cluster-devel <cluster-devel@redhat.com>,
	Christoph Hellwig <hch@lst.de>,
	Linux FS-devel Mailing List <linux-fsdevel@vger.kernel.org>
Subject: Re: [PATCH v8 06/10] iomap: Add page_write_end iomap hook
Date: Tue, 5 Jun 2018 14:17:38 +0200	[thread overview]
Message-ID: <CAHpGcMJGf49BN=bSz26x3wK6+nJT_Nf5qZme1GvaWkxyDqx7=Q@mail.gmail.com> (raw)
In-Reply-To: <20180605120733.boqsxhku77warm7s@twin.jikos.cz>

2018-06-05 14:07 GMT+02:00 David Sterba <dsterba@suse.cz>:
> On Mon, Jun 04, 2018 at 09:31:19PM +0200, Andreas Gruenbacher wrote:
>> --- a/fs/iomap.c
>> +++ b/fs/iomap.c
>> @@ -181,16 +181,22 @@ iomap_write_begin(struct inode *inode, loff_t pos, unsigned len, unsigned flags,
>>
>>  static int
>>  iomap_write_end(struct inode *inode, loff_t pos, unsigned len,
>> -             unsigned copied, struct page *page, struct iomap *iomap)
>> +             unsigned copied, struct page *page, struct iomap *iomap,
>> +             const struct iomap_ops *ops)
>>  {
>> +     typeof(ops->page_write_end) page_write_end = ops->page_write_end;
>
> Is the reason to use typeof is to avoid repeating the type of
> page_write_end?

Yes, the type is void (*)(struct inode *, loff_t, unsigned, struct
page *, struct iomap *), which is a bit bulky.

> As it's only for a temporary variable with 2 uses,
> ops->page_write_end does not hurt readability nor is too much typing.
> I would not recommend using typeof outside of the justified contexts
> like macros or without a good reason.

Andreas