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]
Date:	Thu, 3 Aug 2006 17:15:47 -0700
From:	"Nate Diller" <nate.diller@...il.com>
To:	"Adrian Bunk" <bunk@...sta.de>
Cc:	"Andrew Morton" <akpm@...l.org>, "Jens Axboe" <axboe@...e.de>,
	"Linux Kernel Mailing List" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH -mm] [1/2] Remove Deadline I/O scheduler

On 8/3/06, Adrian Bunk <bunk@...sta.de> wrote:
> On Thu, Aug 03, 2006 at 03:57:32PM -0700, Nate Diller wrote:
>
> > This patch removes the Deadline I/O scheduler.  Performance-wise, it
> > should be superceeded by the Elevator I/O scheduler in the following
> > patch.  I would be very ineterested in hearing about any workloads or
> > benchmarks where Deadline is a substantial improvement over Elevator,
> > in throughput, fairness, latency, anything.
> >...
>
> You are starting with the last step.

You're right, I should have made myself clear.  My goal is not to get
deadline removed, but a discussion with Andrew some months ago showed
he was averse to creating more options than we already have.  So since
I expect elevator can surpass deadline, I wanted to show that I think
deadline is the one that it should replace.  Certainly, CFQ and as can
both beat elevator for a good number of workloads.
>
> First, get your Elevator I/O scheduler reviewed [1] and show some data
> that backs your "it should be superceeded by the Elevator I/O scheduler"
> claim.
>
> Then get your Elevator I/O scheduler included in Linus' tree.

My first priority is to get that patch in order.

>
> Then you might perhaps schedule the Deadline I/O scheduler for removal.

what are people's thoughts on this?  since schedulers are modular, do
we need a scheduled removal, or can this just sit in -mm for a while?
if people are concerned about scripts which ask for 'deadline', we
could add another exception (like the as->anticipatory one).

NATE
-
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/

Powered by blists - more mailing lists