linux-kernel - Re: [PATCH v2 3/4] block: loop: introduce 'use

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

Message-ID: <CACVXFVOgMvccTUZB91OZFgjc_zQk0VnhxtBTO0OhwRYBye8Xnw@mail.gmail.com>
Date:	Tue, 27 Jan 2015 13:26:54 +0800
From:	Ming Lei <ming.lei@...onical.com>
To:	Christoph Hellwig <hch@...radead.org>
Cc:	linux-kernel@...r.kernel.org,
	Dave Kleikamp <dave.kleikamp@...cle.com>,
	Jens Axboe <axboe@...nel.dk>, Zach Brown <zab@...bo.net>,
	Maxim Patlasov <mpatlasov@...allels.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Alexander Viro <viro@...iv.linux.org.uk>,
	Benjamin LaHaise <bcrl@...ck.org>,
	Jeff Moyer <jmoyer@...hat.com>
Subject: Re: [PATCH v2 3/4] block: loop: introduce 'use_aio' sysfs file

On 1/25/15, Christoph Hellwig <hch@...radead.org> wrote:
> On Tue, Jan 13, 2015 at 11:44:47PM +0800, Ming Lei wrote:
>> So that users can control if kernel aio is used to submit I/O.
>
> Why do you want a sysfs flag for this instead of a flags in ->lo_flags
> at setup time?  That's how all other loop-related ABI flags work.

The flag only effects performance or CPU/memory resource utilization,
and I plan to change it as dio flag.

On Tue, Jan 27, 2015 at 1:57 AM, Jeff Moyer <jmoyer@...hat.com> wrote:
> Ming Lei <ming.lei@...onical.com> writes:
>
>> So that users can control if kernel aio is used to submit I/O.
>
> How would a user know to choose aio or, um, /not/ aio?  At the very
> least, documentation is required for this.  I'd rather see the option
> disappear completely, though.

As the comment says in patch 4/4, using direct I/O can save memory/CPU
a lot, and won't hurt throughput if I/O is from filesystem. Otherwise,
user may choose to not use direct I/O for sake of throughput.


Thanks,
Ming Lei
--
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/