[aklug] Re: Phoronix recently published an article regarding a ~200 lines Linux Kernel patch

From: Jim Gribbin <jimgribbin@gmail.com>
Date: Fri Nov 19 2010 - 14:02:40 AKST

And it seems to get more heated later in the discussion.

http://lkml.org/lkml/2010/11/16/392

And there seems to be no clear winner. At least, not one I feel qualified to call.

I think I.m going to go ahead and play around with it a little.

--
Jim Gribbin
Linux user #179129
----- Original Message -----
From: "Bruce Hill" <bruce@slackwarebox.com>
To: "Alaska Linux Users Group" <aklug@aklug.org>
Sent: Friday, November 19, 2010 1:07:35 AM
Subject: [aklug] Re: Phoronix recently published an article regarding a ~200 lines Linux Kernel patch
On Fri, Nov 19, 2010 at 12:49:17AM -0900, Jim Gribbin wrote:
> ...hat improves responsiveness under system strain.
> 
> anybody see this?
> 
> http://www.webupd8.org/2010/11/alternative-to-200-lines-kernel-patch.html
> 
> Tried it on my laptop, seems to make a difference.
> 
> I didn't have any actual benchmarks to try out, but I've been doing a little experimenting w/ video recording off a usb webcam lately. Actually recording video REALLY seems to take quite a bit of horsepower. The way my laptop was, I was to the point of trying to figure out recording from the command line and turning off X for the process.
> 
> I recently upgraded my laptop to a 2.4 GHz P4-M from the original 1.6 GHz. Under Cheese with the 1.6, I was getting about a 20 sec. freeze frame at 128 x 96. With the 2.4, I was getting iffy 320 x 240. With this patch alternative, it's getting pretty acceptable (noticeably better) 320 x 240.
> 
> The guy who worked this out works for RedHat and I'm guessing is using RH Enterprise Desktop. Fedora seems a little different in some of the directory structure. I used exactly the same procedure as recommended for Ubuntu.
> 
> --
> Jim Gribbin
I have used the kernel patch but not the hack from the Lennart (the RH
guy). It is being highly debated on LKML, and all the kernel devels that I
know personally are going with Mike Galbraith's kernel patch versus the
hack from Lennart Poettering -- including Linus. Here's the LKML thread
with the original info about the patch:
http://lkml.org/lkml/2010/10/19/123
where I noticed a while back. I have patched it against 2.6.36 and am now
running it on 2.6.37-rc2.
If you go look at Hot Messages on LKML you'll get an eye full.
This is the type of thing that comes around ever so often, and why I am
always testing new git kernels. ;)  Over the years some pretty interesting
changes have ocurred, and I suspect when 'the dust settles' on this issue,
we are going to see some serious performance gains integrated into the
mainline kernel; not some spurious hack such as CK patches have been over
the years. ;)
Just my 2c
Bruce
-- 
May the Lamb that was slain receive the reward of His suffering!
"Experience is a hard teacher because she gives the test first, the
lesson afterward. But properly learned, the lesson forever changes
the man."
---------
To unsubscribe, send email to <aklug-request@aklug.org>
with 'unsubscribe' in the message body.
---------
To unsubscribe, send email to <aklug-request@aklug.org>
with 'unsubscribe' in the message body.
Received on Fri Nov 19 14:02:52 2010

This archive was generated by hypermail 2.1.8 : Fri Nov 19 2010 - 14:02:52 AKST