A running description of activity related to DragonFly BSD.
Hiten Pandya brought up futexes for possible inclusion in DragonFly, pointing at this page for reference.
Bill Huey brought up http://people.redhat.com/drepper/futex.pdf, with the caveat that futexes are used for 1:1 threading models, not M:N models such as DragonFly is using/will be using.
Jeroen Ruigrok posted a whole pile of links to routing information for anyone wanting to work on it:
Basics:
TCP Vegas: End to End Congestion Avoidance on a Global Internet
http://citeseer.nj.nec.com/brakmo95tcp.html
Evaluation of TCP Vegas: Emulation and Experiment
http://citeseer.nj.nec.com/ahn95evaluation.html
TCP and Successive Fast Retransmits
http://citeseer.nj.nec.com/floyd95tcp.html
Congestion Avoidance and Control
http://citeseer.nj.nec.com/jacobson88congestion.html
Performance Problems in BSD4.4 TCP
http://citeseer.nj.nec.com/brakmo95performance.html
Comparison of Tahoe, Reno, and SACK
http://citeseer.nj.nec.com/fall95comparisons.html
TCP Vegas: New Techniques for Congestion Detection and Avoidance
http://citeseer.nj.nec.com/brakmo94tcp.html
A Tree-based Packet Routing Table for Berkeley Unix
http://www.cs.berkeley.edu/~sklower/routing.ps
Newer:
TCP Vegas Revisited
http://citeseer.nj.nec.com/hengartner00tcp.html
Analysis on TCP Vegas and TCP Reno
http://citeseer.nj.nec.com/hellal00analysis.html
Scalable Timers for Soft State Protocols
http://citeseer.nj.nec.com/48810.html
End-to-End Internet Packet Dynamics
http://citeseer.nj.nec.com/paxson97endtoend.html
End-to-End Routing Behavior in the Internet
http://citeseer.nj.nec.com/3573.html
Routing with a Clue
http://citeseer.nj.nec.com/afek99routing.html
IP Lookups using Multiway and Multicolumn Search
http://citeseer.nj.nec.com/lampson98ip.html
Fast and Scalable Layer Four Switching
http://citeseer.nj.nec.com/srinivasan98fast.html
High-Speed Policy-based Packet Forwarding Using Efficient
Multi-dimensional Range Matching
http://citeseer.nj.nec.com/lakshman98highspeed.html
Fast Address Lookups using Controlled Prefix Expansion
http://citeseer.nj.nec.com/srinivasan99fast.html
Matt Dillon’s made a diary update.
The memory slab allocator is now on by default, and the old system is gone.
Man pages for pim(4) and multicast(4) for “Protocol Independent Multicasting”, are in.
Sysinstall has been modified with changes for NFS; you can pick NFSv3 (default) and TCP protocol (not default), now, along with some internal changes.
The SysV IPC regression test suite has been added, in src/tools/regression/
David P. Reese posted his 9th step into syscall separation, and noted that there remains:
Kip Macy’s checkpointing code has been committed; I’m pasting Matt Dillon’s post about it as there’s a lot of issues to consider.
For those of you late to the party, checkpointing allows you to “freeze” a copy of an application so that, in theory, you can restore the program to that running state at a later point in time. Useful, for instance, if you have a program that takes a long time to complete and you don’t want to have to restart from the beginning if there’s an interruption.
Continue reading “Checkpointing is in”
David Cuthbert brought up the idea of Doxygen headers for the source; Matt Dillon didn’t care for the idea, pointing at this for an example. Hiten Pandya noted that he’d like to have a separate by-hand handbook, for which Eirik Nygaard posted a possible example. Matthew Fuller added that he is working on a SGML application for library documentation as part of another project.
While noting that he has made a number of changes to the scheduler, Matt Dillon described a new tool called ‘wmake’. wmake allows you to run make in a subdirectory simulating a buildworld environment, without actually having to build world.
(quoted from his post, using libkvm for an example:)
cd /usr/lib/libkvm
wmake obj
wmake
make install
Joshua Coombs has an extensive writeup of his ideas for routing, located at http://www.outofspec.com/routing/
While talking about something else, Hiten Pandya mentioned the steps necessary to turn on profiling, to help catch bugs. I’ve saved the steps here, in case they are useful…
Continue reading “How to profile”
Emiel Kollof noted that it would be nice if the splash screen loaders could read gzipped files, so that any splash screens could be stored in a compressed format and still used. Hiten Pandya pointed at kern/imgact_gzip.c and boot/i386/kgzldr/boot.c for examples, if anyone wants to tackle this project.
/bin/sh had a big introduced recently that would make it crash when booting with a new world. It’s been fixed, so recompile /bin/sh with updated sources when possible.
David Leimbach noted that Ron Minnich was porting Plan 9 namespaces to FreeBSD, which duplicates some of the security features to be covered by VFS. Ron Minnich’s web page has more data on this and other technologies.
usr/share/examples/splash has been committed. This is the logo screen Emiel Kollof made from the DragonFly logo.
Emiel Kollof’s also got the NVIDIA binary video driver working; it should show up in dfports.
Hiten Pandya (hmp@backplane.com) is looking for anyone using a AMI MegaRAID card, who is willing to test a patch for FreeBSD-4/DragonFly.
dfports/emulators/vmware2 is in, though Matt Dillon warns that it is not yet functional.
File System Exerciser, in tools/regression/fsx, has been added, by David Rhodus.
fdisk now has a -p option, to let it operate on disk images.
David Rhodus has fixed the problems reported by Richard Sharpe where Linux NFS clients couldn’t achieve above 5 MBps throughput.
I wrote down the explicit steps I used to get a DragonFly machine running and up-to-date; I’m including them here for the benefit of others. This was working as of mid-September.
** Updated December 2003; silly typos.
** Update February 2004: This guide was written using FreeBSD 4.8 as a guide; it may break. There is a guide page on dragonflybsd.org that talks about upgrading from FreeBSD 4.9.
Continue reading “Installing from source”
