LWN: Comments on "Trip report: the Ottawa Linux Symposium"

RE: Rik van Riel's Talk

Zakaelri — Sat, 11 Oct 2003 17:32:28 +0000

Someone needs to write a "Kernel Hacking for Dummies" book to help out us application programmers. :-) I've picked up bits and pieces on the net, but some of them are probably assembled in the wrong order. How in the world do you learn this stuff?

Well, simple google search got me these few links on Kernel Hacking:
http://www.kernelhacking.org/
http://www.linuxchix.org/content/courses/kernel_hacking/
And I could swear there was an O'Reilly book on it too... but I can't find the listing on their site.

I would assume that many of the people who do kernel hacking have been in the field for quite some time, and that they have learned about kernel hacking from either college or a *nix-related job somewhere around the line.

Why doesn't my laptop suspend?

otterley — Fri, 08 Aug 2003 22:24:11 +0000

Lack of consistent suspend/resume features isn't the only thing missing from Linux laptop support. Also missing is the ability to adequately deal with docking stations. Windows has the capability to change hardware profiles -- not quite on-the-fly, but definitely across a suspend/resume cycle -- and Linux has no such capability at present.

Lots of user-level support will be needed to handle this too. For example, the window system will have to handle a display connection and disconnection -- when the laptop is docked, it will likely be plugged into a monitor with a higher resolution and different H/V frequency ranges. All without shutting down and restarting the X server.

As you can see, we have a long way to go.

Keynote talk

guest1 — Thu, 07 Aug 2003 16:23:39 +0000

http://sourceforge.net/projects/tdb/

Keynote talk

cspalletta — Thu, 07 Aug 2003 13:16:30 +0000

On slide 4 from the OLS Keynote Address, Rusty Russell says:

> Read the Original TDB Source code..

It seems to me I should know what TDB is but somehow I can't remember.
So, what is TDB? Could some kind person refresh my memory

Keynote talk

Lavonardo — Thu, 07 Aug 2003 08:09:48 +0000

Here's the URL for Rusty's slides: http://ozlabs.org/~rusty/ols-2003-keynote/ols-keynote-2003.html.

Keynote talk

daniel — Mon, 04 Aug 2003 23:00:39 +0000

"Are there others people find valuable?"

zinf

MP3s of the speeches

ebresie — Mon, 04 Aug 2003 20:41:00 +0000

I like the proceedings provided for this iteration of the symposium, but was curious if there will be mp3 versions of the speeches like what seems to exist for past years.

Keynote talk

dac — Fri, 01 Aug 2003 12:50:12 +0000

Thanks.

Are there others people find valuable?

Keynote talk

yzhuang — Fri, 01 Aug 2003 09:22:43 +0000

gperf
qemu
rsync
valgrind
ccache
distcc

I guess it is Rusty's personal interest...

Keynote talk

mulix — Thu, 31 Jul 2003 21:03:25 +0000

valgrind
ccache
distcc
gprof

and two others I don't recall.

Keynote talk

dac — Thu, 31 Jul 2003 17:37:08 +0000

I thought Rusty's keynote address was excellent. Does anyone remember the 5 or 6 applications that he said were key to kernel development? Unfortunately his talk is not in the proceedings.

Rik van Riel's Talk

stuart2048 — Wed, 30 Jul 2003 20:06:06 +0000

For years I have dreamt of a Java environment that, instead of
relying on the OS to blindly manage pages, integrates the paging
system directly into the object heap. So that the Java VM pages
objects, or parts of objects that aren't being used (but are
not quite garbage yet).

Of course I expect some of you will tell me this was implemented
30 years ago in some other language ;-)

Link to Rik van Riel's paper

tjc — Tue, 29 Jul 2003 15:08:36 +0000

Thank you!

Rik van Riel's Talk

dancres — Tue, 29 Jul 2003 08:49:53 +0000

I tend to agree but the number of objects traversed at any point is also dependent on the algorithm chosen. Certain types of GC will be far worse in respect of this behaviour than others.

I'd sure be interested in hearing Rik "name and shame" the various VM implementations.

pjdc — Tue, 29 Jul 2003 08:26:42 +0000

The paper accompanying Rik's talk is here.

Rik van Riel's Talk

walters — Tue, 29 Jul 2003 06:08:41 +0000

Does the JVM walk through it's entire virtual address space, or does it just visit pages that have been marked in some way? The former would be horrible, but I'm not sure the later is possible on most architectures.

Neither. Most of them only traverse allocated objects, looking at pointers. But I think what Rik is talking about is that this may end up hitting the majority of pages allocated. See any paper on garbage collection for more information.

Rik van Riel's Talk

Webexcess — Tue, 29 Jul 2003 04:19:55 +0000

Does the JVM walk through it's entire virtual address space, or does it just visit pages that have been marked in some way?

I don't know enough about Java VMs to say. :-|

By "bad for the elevator", you are talking about the disk allocation algorithm?

The elevator code is there to reduce disk seeking by reordering disk operations. Unfortunately, when one operation depends on the preceeding one, it never gets a chance. If your page tables are on disk, then you don't know where on disk the page data you need is until after you've fetched the tables. Ie, paging gets alot slower.

Rik van Riel's Talk

tjc — Tue, 29 Jul 2003 03:08:34 +0000

An example mentioned was how some Java virtual machines that do garbage collection "at some point in the future" tend to crawl through all available memory object allocation by object allocation.

Also the current page table is growing too big, and needs to be replaced.

Isn't this tied to the hardware architecture, more or less?

(bad for the elevator)

By "bad for the elevator", you are talking about the disk allocation algorithm?

Rik van Riel's Talk

Webexcess — Tue, 29 Jul 2003 01:55:50 +0000

One thing he talked about was how hard it is to deal with programs that behave badly. An example mentioned was how some Java virtual machines that do garbage collection "at some point in the future" tend to crawl through all available memory object allocation by object allocation. This puts pressure on the memory of other applications that haven't been used recently, like say your mail app. When you try to switch back to that other app you find that it's been completely paged out (chug.. chug.. chug..).

Also the current page table is growing too big, and needs to be replaced. The obvious options include dynamic page table allocation and swapping out page tables. Neither of these options is good - the first imposes a general system-wide slowdown on page allocation, the second could have horrible impact when you swap some page tables in only to realize that you need to swap the pages themselves in too (bad for the elevator).

Rik van Riel's Talk

tjc — Tue, 29 Jul 2003 01:26:14 +0000

Can anyone tell me more about Rik van Riel's talk? I'm interested in what he had to say about page replacement algorithms in particular.