LWN: Comments on "Kernel development statistics for 2.6.35"

Kernel development statistics for 2.6.35

abadidea — Thu, 22 Jul 2010 14:05:17 +0000

I am guessing they have a separate budget, mission, and of course branding from the parent Intel. They have an established reputation so it makes sense IMHO to list them separately.

Canonical?

hppnq — Mon, 19 Jul 2010 07:11:35 +0000

I don't know how I missed that discussion originally.

Wow. You didn't.

Canonical?

mikov — Sun, 18 Jul 2010 22:24:32 +0000

Wow, that is a can of worms! :-) I don't know how I missed that discussion originally. I see that not much has changed since 2008 in that respect.

Canonical?

rahulsundaram — Sun, 18 Jul 2010 21:33:17 +0000

That was covered and debated at

http://lwn.net/Articles/298864/

Canonical?

mikov — Sun, 18 Jul 2010 21:12:26 +0000

Is Canonical involved at all in kernel development? I see Mandriva, who is supposed to be almost dead, but Canonical is missing completely.

Kernel development statistics for 2.6.35

martinfick — Fri, 16 Jul 2010 19:58:15 +0000

What would be interesting would be to see these number for other OSs! Would they be much higher by virtue of more drivers in linux (perhaps not than windows?). Would the filesystems be proportionally way less on other OSes also contributing to a "smaller" (not really, just %) core on linux?

Kernel development statistics for 2.6.35

jake — Fri, 16 Jul 2010 04:46:16 +0000

> I hope the text can be amended so my name is spelled correctly for my
> brief moment of fame :)

Our apologies! Fixed now.

jake

Kernel development statistics for 2.6.35

gregkh — Fri, 16 Jul 2010 04:33:11 +0000

Ick, so sorry about that. Hopefully the editor can make that change.

Again, my apologies.

Kernel development statistics for 2.6.35

Eliot — Fri, 16 Jul 2010 04:24:17 +0000

I hope the text can be amended so my name is spelled correctly for my brief moment of fame :)

"The company AudioScience Inc. sneaks onto the list of changes due to all of the work that Eliot Blennerhassett has been doing on the asihpi sound driver."

Kernel development statistics for 2.6.35

Julie — Thu, 15 Jul 2010 19:44:35 +0000

Let's try something different this time, and break the contributions down by the different functional areas of the kernel.

Presenting the material in this way is really interesting and illuminating. I'd often frequently been intrigued when trawling through the source tree by how little proportionally the 'core' code takes up, but still, only 5% surprised me.

Kernel development statistics for 2.6.35

cesarb — Thu, 15 Jul 2010 14:16:44 +0000

Well, it depends on how you parse that sentence. You *did* delete almost all of them, leaving only a few lines on each. That is, you deleted almost all of the content of the files.

Kernel development statistics for 2.6.35

arjan — Thu, 15 Jul 2010 14:06:42 +0000

Interesting that Wind River is counted separately from Intel.... Intel acquired Wind River like a year ago or so ....

Kernel development statistics for 2.6.35

ukleinek — Thu, 15 Jul 2010 12:22:24 +0000

"deleted almost all of the ARM default config files" is wrong. I didn't delete a single defconfig, just removed ~1000 lines in average from each defconfig.

Thanks
Uwe

Ceph and Dreamhost

Cato — Thu, 15 Jul 2010 11:05:18 +0000

As a Dreamhost hosting customer, I'm rather pleased to see they are involved in Ceph filesystem development (via their co-founder) and are apparently <a href="http://www.dreamhost.com/jobs.html">hiring Ceph developers</a>.

On a more prosaic note: not sure if Ceph addresses this, but anything that can fix the multi-hour fsck's required (for shared hosting and VPS providers on ext3) after a power cut would be very beneficial.

At least <a href="http://thunk.org/tytso/blog/2008/08/08/fast-ext4-fsck-times/">ext4 fsck is faster</a>...

Kernel development statistics for 2.6.35

dvhart — Thu, 15 Jul 2010 02:31:21 +0000

Great breakdown Greg, thanks for doing this. Very informative.