Getting started on a Summer of Code project

If you were thinking of working on a disk scheduler for DragonFly, this is your lucky day!  Brills Peng asked for some overall guidance on how to start on a Summer of Code project.  I threw out some general tips, Alex Hornung talked up resources on kernel programming, and Venkatesh Srinivas described exactly what you’d need to write a disk scheduler.  There’s about 50% of a whole proposal, prewritten.

Summer of Code 2011: We’re in!

We made it into Google Summer of Code for a 4th year!  (yay!)

http://www.google-melange.com/gsoc/org/show/google/gsoc2011/dragonflybsd

If you want to mentor, apply here:

http://www.google-melange.com/gsoc/mentor/request/google/gsoc2011/dragonflybsd

(You will need to create a login if you don’t have one.)  I’m assuming the applicants are going to be people I know with a direct history with DragonFly; otherwise be prepared to give a good history.  Signing up to mentor does not mean you must mentor if there aren’t any projects that interest you; it does mean you need to review applications and provide feedback for students March 28th – April 8th.

If you want to be a student with DragonFly:

Check the projects page for ideas:
http://www.dragonflybsd.org/docs/developer/gsocprojectspage/

… or come up with your own.

Get your application together by March 28th.  Start talking about it on the mailing list or IRC or however as soon as you can; there’s a direct relationship between the amount of preparation we see beforehand and people getting accepted.

Here’s the timeline:

http://www.google-melange.com/document/show/gsoc_program/google/gsoc2011/timeline

Copied from my email to users@/kernel@, cause it has everything you need.

 

 

More microbenchmarks

Venkatesh Srinivas performed the fefe.de ‘scalability’ benchmarks, which have been mentioned here before.  He performed it on different hardware and only with DragonFly, so it’s not really for comparison but rather for analysis.  However: graphs!

Matthew Dillon added some system tunables to match these microbenchmarks, and then changed the values.  The benchmarks looked better, but according to him you wouldn’t want to run a system normally with those values.

pkgsrc current and 2.9, plus shallow pulls

There’s two recent changes for pkgsrc and DragonFly:

BSD Needs Books, the video

Michael Lucas’s “BSD Needs Books” talk from NYCBSDCon 2010 is online, in video form.  I got to see this as it happened, and it was a excellent talk.  Mr. Lucas is able to put some reasonable arguments together as to the why of things, since he’s been published multiple times, plus his sense of humor keeps it moving.

Hey, wait – there’s more from the conference on BSD TV!  How did I miss this?  Hopefully even more will show up; the facility was perfect for recording.

The best way to fix up pkgsrc…

… is to make its patches unnecessary, by getting the changes needed for any program to compile on DragonFly built right into the program.  (Often called “pushing patches upstream”)  That usually means creating a patch and then tracking down the program authors to get them to include those changes in the next release of a project.  That tracking down can be a majority of the work.  In that case: thanks, Rumko!

Update: Also, thanks, Matthias Rampke!  He did the same thing for pcc.