Posted on
Holy crap, look at those numbers

Remember the new scheduler work?  Well, it continued, and now Francois Tigeot has posted pgbench benchmarks of the progress and benchmarks of DragonFly vs. other operating systems.  The links are to PDFs; scroll down as each have multiple pages.

The summary result: If you’re running Postgres, you probably want to do it on DragonFly.  The numbers are the best results for any BSD,  even better to some extent than Linux, which has had its own issues with schedulers and Postgres.  DragonFly 3.2 will include these improvements.

Posted on
3.2 branch and release plan

As I typed elsewhere, my general plan is to branch DragonFly 3.2 on the 8th, and release on the 22nd.  That should give the recent scheduler and gcc work a chance to settle, and perhaps get a new version of USB support in too.  It will probably be using pkgsrc-2012Q3, also, though we may not have binary i386 packages.  3.2 is shaping up to be a much more significant release than I expected.

 

Posted on
gcc 4.7 in, gcc 4.1 out

John Marino has accomplished the difficult task of putting gcc 4.7 into DragonFly.  Version 4.4 is still the default, and the older 4.1 version has been disabled.  If you want to try this newer version, setting WORLD_CCVER=gcc47 will build kernel and world that way too.  If you’re curious about what’s different in this version of gcc, there’s a 4.7 changelog.

Are we the only BSD with this new a version in base?  I think so.

P.S.: You’ll want to do a full buildworld if you’re running DragonFly 3.1

P.P.S.: you may need to put ‘NO_GCC47=true’ in make.conf, going from IRC comments.

P.P.P.S.: Nope, now it’s fine.

 

Posted on
Lazy Reading for 2012/09/30

It’s been an extremely busy week for me, but I still have a batch of links here.

Your unrelated link of the week: Did you know one of the original ideas was to name DragonFly “TortoiseBSD” “TurtleBSD”?  Probably not the best name.

Posted on
Posting but not reading mailing lists

The old mailing list software for @dragonflybsd.org mailing lists, bestserv, apparently allowed people not subscribed to a list to post to it, after answering a confirmation message for each message posted.

The closest way to duplicate that for Mailman is to sign up for the list you want, and then turn off mail delivery for your email address in the config page for that mailing list.  This won’t affect a lot of people, since most people want list output in their mailbox, but there’s at least a few I’ve fixed that way.

Posted on
A flurry of fixes and scheduler improvements

The combination of Mihai Carabas’s successful Summer of Code work on the scheduler and the recent Postgres benchmarking got Matthew Dillon to start thinking about making UNIX domain sockets work better, a shortcut around the buffer cache, scheduler improvements and then a new default scheduler, along with a change in idle CPU behavior.  The best place to understand all the changes is in his long post to users@.

We should have benchmarks soon to show the performance improvements from all this.