linux-kernel - Re: [PATCH -v2 4/6] fuse: add a sysfs parameter to control the maximum request size

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Wed, 08 Aug 2012 16:43:54 +0200
From:	Miklos Szeredi <miklos@...redi.hu>
To:	Mitsuo Hayasaka <mitsuo.hayasaka.hu@...achi.com>
Cc:	Alexander Viro <viro@...iv.linux.org.uk>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Muthukumar R <muthur@...il.com>,
	fuse-devel@...ts.sourceforge.net, linux-kernel@...r.kernel.org,
	linux-fsdevel@...r.kernel.org, linux-doc@...r.kernel.org,
	yrl.pp-manager.tt@...achi.com, Nikolaus Rath <Nikolaus@...h.org>,
	Liu Yuan <namei.unix@...il.com>,
	Has-Wen Nienhuys <hanwen@...all.nl>
Subject: Re: [PATCH -v2 4/6] fuse: add a sysfs parameter to control the maximum request size

Mitsuo Hayasaka <mitsuo.hayasaka.hu@...achi.com> writes:

> Add a max_pages_per_req sysfs paramater to limit the maximum
> read/write request size. It can be changed to arbitrary number
> between 32 and the nr_pages equivalent to pipe_max_size, and the
> 32 pages are set by default.
>
> The sysfs parameter control is required, as follows.
>
> * The libfuse should change the current MIN_BUFSIZE limitation
>   according to the current maximum request size in FUSE. If not,
>   the libfuse must always set MIN_BUFSIZE to the maximum request
>   limit (= [nr_pages (equivalent to pipe_max_size) * 4KB + 0x1000]),
>   which leads to waste of memory.

I don't see the purpose of this sysfs parameter.  Userspace can
calculate the needed buffer size from the max_read/max_write parameters,
can't it?

>
> * It is easy to find and set the paramter to the optimized value
>   in order to improve the read/write throughput, since the
>   maximum request limit does not always provides the highest
>   throughput.

So basically this is a global max_read/max_write limit?  For that a
better solution would be to add it to /etc/fuse.conf.

But I have doubts about whether this is useful.

Thanks,
Miklos
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/