Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 20 Jan 2021 10:12:01 -0500 (EST)
From:   Mikulas Patocka <mpatocka@redhat.com>
To:     Jan Kara <jack@suse.cz>
cc:     Dave Chinner <david@fromorbit.com>,
        Zhongwei Cai <sunrise_l@sjtu.edu.cn>,
        "Theodore Ts'o" <tytso@mit.edu>,
        Matthew Wilcox <willy@infradead.org>,
        David Laight <David.Laight@aculab.com>,
        Mingkai Dong <mingkaidong@gmail.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Steven Whitehouse <swhiteho@redhat.com>,
        Eric Sandeen <esandeen@redhat.com>,
        Dave Chinner <dchinner@redhat.com>,
        Wang Jianchao <jianchao.wan9@gmail.com>,
        Rajesh Tadakamadla <rajesh.tadakamadla@hpe.com>,
        linux-kernel <linux-kernel@vger.kernel.org>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        linux-nvdimm <linux-nvdimm@lists.01.org>
Subject: Re: Expense of read_iter
In-Reply-To: <20210120141834.GA24063@quack2.suse.cz>
Message-ID: <alpine.LRH.2.02.2101200951070.24430@file01.intranet.prod.int.rdu2.redhat.com>
References: <alpine.LRH.2.02.2101061245100.30542@file01.intranet.prod.int.rdu2.redhat.com> <20210107151125.GB5270@casper.infradead.org> <17045315-CC1F-4165-B8E3-BA55DD16D46B@gmail.com> <2041983017.5681521.1610459100858.JavaMail.zimbra@sjtu.edu.cn>
 <alpine.LRH.2.02.2101131008530.27448@file01.intranet.prod.int.rdu2.redhat.com> <1224425872.715547.1610703643424.JavaMail.zimbra@sjtu.edu.cn> <20210120044700.GA4626@dread.disaster.area> <20210120141834.GA24063@quack2.suse.cz>
User-Agent: Alpine 2.02 (LRH 1266 2009-07-14)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk


On Wed, 20 Jan 2021, Jan Kara wrote:

> Yeah, I agree. I'm against ext4 private solution for this read problem. And
> I'm also against duplicating ->read_iter functionatily in ->read handler.
> The maintenance burden of this code duplication is IMHO just too big. We
> rather need to improve the generic code so that the fast path is faster.
> And every filesystem will benefit because this is not ext4 specific
> problem.
> 
> 								Honza

Do you have some idea how to optimize the generic code that calls 
->read_iter?

vfs_read calls ->read if it is present. If not, it calls new_sync_read. 
new_sync_read's frame size is 128 bytes - it holds the structures iovec, 
kiocb and iov_iter. new_sync_read calls ->read_iter.

I have found out that the cost of calling new_sync_read is 3.3%, Zhongwei 
found out 3.9%. (the benchmark repeatedy reads the same 4k page)

I don't see any way how to optimize new_sync_read or how to reduce its 
frame size. Do you?

Mikulas