[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <0618f315-7061-c3fd-15d3-c19cea48cc4c@gmail.com>
Date: Sat, 1 Aug 2020 13:12:22 +0300
From: Pavel Begunkov <asml.silence@...il.com>
To: Christoph Hellwig <hch@...radead.org>
Cc: Alexander Viro <viro@...iv.linux.org.uk>,
linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] splice: direct call for default_file_splice*()
On 30/01/2020 19:54, Christoph Hellwig wrote:
> On Mon, Jan 20, 2020 at 11:49:46PM +0300, Pavel Begunkov wrote:
>> Indirect calls could be very expensive nowadays, so try to use direct calls
>> whenever possible.
Hah, I'm surprised to find it as
00c285d0d0fe4 ("fs: simplify do_splice_from").
Christoph, even though this one is not a big deal, I'm finding the
practice of taking others patches and silently sending them as yours
own in general disgusting. Just for you to know.
>
> ... and independent of that your new version is much shorter and easier
> to read. But it could be improved a tiny little bit further:
>
>> if (out->f_op->splice_write)
>> - splice_write = out->f_op->splice_write;
>> + return out->f_op->splice_write(pipe, out, ppos, len, flags);
>> else
>> - splice_write = default_file_splice_write;
>> -
>> - return splice_write(pipe, out, ppos, len, flags);
>> + return default_file_splice_write(pipe, out, ppos, len, flags);
>
> No need for the else after an return.
>
>> if (in->f_op->splice_read)
>> - splice_read = in->f_op->splice_read;
>> + return in->f_op->splice_read(in, ppos, pipe, len, flags);
>> else
>> - splice_read = default_file_splice_read;
>> -
>> - return splice_read(in, ppos, pipe, len, flags);
>> + return default_file_splice_read(in, ppos, pipe, len, flags);
>
> Same here.
>
--
Pavel Begunkov
Powered by blists - more mailing lists