Alternative Performance Kernel for Debian

[gnudude]

Debian Sid user here

sidux isnt sid....

[Soul Singin']

As for all the attacks on damentz, I feel they are totally unwarranted and the admin agrees with this.

Of course it LOOKS unwarranted.

He removed the objectionable material and then he used a hot-linked image to insert profanity into my post:

• http://liquorix.net/imgs/fat_lines_light-dmz.png

EDIT and UPDATE: And then he switched the image back to cover his tracks.
.

[damentz]

Debian Sid user here

sidux isnt sid....

sidux is Debian sid with a better (actually questionable according to what you prefer) default configuration and package selection with the desktop user in mind. The developers also try to temporarily fix packages that arrive in debian Sid which might have broken many of their user's installations.

What you asserted it similar to saying, Linux Mint is not Ubuntu, although with a few commands with apt-get, and a few gtkrc theme modifications, you will be using stock Ubuntu.

The trick is that both of those distributions depend highly on their distribution base's repository always functioning and doing the work for them - they just change some cosmetics, default configuration, and default package selection.

[jalu]

sidux socks

[gnudude]

hey if you want to dance off topic then lets do a jig...

sidux is Debian sid

uh no....sidux doesn't even use a debian kernel but rather their own kernel...and you aren't suppose to use some tools that debian specifically suggests to use...no dfsg...etc....

sidux is based on debian sid....sidux is debian sid based...sidux is binary compatible with sid....but sidux is not sid....

The developers also try to temporarily fix packages that arrive in debian Sid

that would make it NOT sid then...right...

similar to saying, Linux Mint is not Ubuntu

actually clem uses a number of specific tools that he has wrote for mint....but who is talking about mint...

I guess linspire is also debian sid....xandros.....ubuntu....all debian sid....and...well....everything is debian sid....glad we cleared that up...

[damentz]

Hmm, not necessarily, because it's easy to remove all the sidux cruft and turn it into pure debian sid with _just_ apt-get. Yes, it's built on top of debian Sid, but it's essentially Debian Sid with certain changes that are easy to revert.

As far as the kernel, it's not necessarily in everyone's best interest (mine at least), so I remove that and put my own. If I run Debian Sid with the Liquorix kernel, am I running Debian Sid anymore, or is that just another package discrepancy from vanilla sid?

[gnudude]

well if sidux is sid....then I will go over to the sidux forums and ask about my sid install since if sidux is sid then sid must be sidux as well...

[AdrianTM]

If sidux is sid then why does it exists? Since we already have a sid...

[gnudude]

..so some can say they are running sid when they are actually running sidux (sid with training wheels)

[Lavene]

This is the "howto" section people... how about trying to stay on topic? Feel free to take this <sarcasm> new and oh so exciting </sarcasm> discussion to the OT or General section...

[craigevil]

All I can say is the liquorix kernel is much more responsive on my crappy system than the debian kernel.

$ inxi -F
System: Host craigevil Kernel 2.6.31-1.dmz.1-liquorix-686 i686 (32 bit) Distro Debian GNU/Linux squeeze/sid
CPU: Single core Intel Pentium 4 (UP) cache 1024 KB flags (sse3 nx lm) bmips 5593.51
Clock Speeds: (1) 2793.104 MHz (2) 2793.104 MHz
Graphics: Card Intel 82915G/GV/910GL Integrated Graphics Controller X.Org 1.6.3.901 Res: 1280x1024@60.0hz
GLX Renderer Mesa DRI Intel 915G GEM 20090712 2009Q2 RC3 x86/MMX/SSE2 GLX Version 1.4 Mesa 7.5.1 Direct Rendering Yes
Audio: Card Intel 82801FB/FBM/FR/FW/FRW (ICH6 Family) High Definition Audio Controller driver HDA Intel
Sound: Advanced Linux Sound Architecture Version 1.0.21
Network: Card Intel 82562ET/EZ/GT/GZ - PRO/100 VE (LOM) Ethernet Controller driver e100 at port dcc0
Disks: HDD Total Size: 160.0GB (9.6% used) 1: /dev/sda WDC WD1600JS-75N 160.0GB
Partition: ID:/ size: 103G used: 15G (15%)
Info: Processes 101 Uptime 5 min Memory 120.6/493.5MB Client Shell inxi 1.1.13

[AdrianTM]

Debian kernel is set up for servers. It boggles the mind why they don't offer a desktop kernel (or am I wrong about this?)

[gnudude]

....much more responsive ....

how so, could you bootchart or time some processes using the debian kernel and then using the liquorix kernel?

[damentz]

Liquorix uses full preemption without preemptable RCU, so it should certainly provide less latencies than debian kernel, which uses no preemption (besides what the kernel does by design to reduce latency).

Debian kernel is set up for servers. It boggles the mind why they don't offer a desktop kernel (or am I wrong about this?)

You're not wrong; either they should balance the kernel configuration in debian to provide a bit less latency at the cost of more overhead or build two kernels that are optimized for their stated purpose.

I would prefer the latter, which is where Liquorix comes in.

[gnudude]

...any numbers to show the improvement in performance?

[damentz]

Code: Select all

liquorix 2.6.31-1.dmz.2
real	0m58.328s
user	1m33.332s
sys	0m14.866s

debian 2.6.30
real	1m5.057s
user	1m31.282s
sys	0m14.525s

The liquorix kernel completes a make -j2 7 seconds before the debian kernel. I ran make allnoconfig on a clean vanilla 2.6.31 kernel for both benchmarks.

To explain why the user time is longer for the liquorix kernel... well I was surfing the web and talking to friends at the same time . The debian 2.6.30 kernel was on a clean boot to the gnome desktop with no arbitrary apps running.

Some definitions, real: total wall clock elapsed time - user: process time outside of kernel - sys: process time making kernel calls.

In this case, sys time is higher for the liquorix kernel is most likely due to full preemption, while debian uses none.

[jalu]

...any numbers to show the improvement in performance?

i wrote:

with liquorix its (boottime) 14 seconds,

my stock debian kernel needs ~30 boottime.
besides that pure numbers it really feels very responsive, fast and stabel.

on another note:
for me its not big fun moving allready compiled kernels around between different hard-disks, qemu and virtualbox.
installing liqourix goes in the blink of an eye.
And i too am not of the ones who think the biggest fun ever is compiling a kernel on your own.

Its good and useful. Thanks
greetings

[Soul Singin']

To explain why the user time is longer for the liquorix kernel... well I was surfing the web and talking to friends at the same time . The debian 2.6.30 kernel was on a clean boot to the gnome desktop with no arbitrary apps running.

In other words, you are so incompetent that you cannot even perform the simplest of all objective comparisons.

Amazing.

I do not know how Jalu timed his boot, but (at the very least) his comparison appears objective. You couldn't do the same?

Let me spell it out for you:

Code: Select all

$ su
Password:
# apt-get update
# apt-get install bootchart bootchart-view
# apt-get clean
# exit

Reboot and append:

Code: Select all

init=/sbin/bootchartd

to GRUB's kernel line.

After you have logged back in, run:

Code: Select all

$ bootchart /var/log/bootchart.tgz

Post the bootchart.png file in this thread, so that we can see the length of time it took to boot your system.

Repeat the process with the other kernel. Then we'll have an objective comparison.
.

[damentz]

To explain why the user time is longer for the liquorix kernel... well I was surfing the web and talking to friends at the same time . The debian 2.6.30 kernel was on a clean boot to the gnome desktop with no arbitrary apps running.

In other words, you are so incompetent that you cannot even perform the simplest of all objective comparisons.

Amazing.

I do not know how Jalu timed his boot, but (at the very least) his comparison appears objective. You couldn't do the same?

Let me spell it out for you:

Code: Select all

$ su
Password:
# apt-get update
# apt-get install bootchart bootchart-view
# apt-get clean
# exit

Reboot and append:

Code: Select all

init=/sbin/bootchartd

to GRUB's kernel line.

After you have logged back in, run:

Code: Select all

$ bootchart /var/log/bootchart.tgz

Post the bootchart.png file in this thread, so that we can see the length of time it took to boot your system.

Repeat the process with the other kernel. Then we'll have an objective comparison.
.

I don't take requests from trolls, do the benchmarks yourself.

[Soul Singin']

do the benchmarks yourself.

How can you claim that your kernel is faster if you haven't made an objective comparison?
.