Slashdot Mirror
Dragonfly BSD 3.2 Released
An anonymous reader writes "Dragonfly BSD recently announced the release of version 3.2 of their operating system. Improvements include: USB4BSD, a second-generation USB stack; merging of a GSoC project to provide CPU topology awareness to the scheduler, giving a nice boost for hyperthreading Intel CPUs; and last but not least, a new largely rewritten scheduler. Some background is in order for the last one. PostgreSQL 9.3 will move
from SysV shared memory to mmap for its shared memory needs. It turned
out that the switch much hurts its performance on the BSDs. Matthew Dillon was fast to respond with a search for bottlenecks and got the performance up to par with Linux."
have they fixed the ssh port bug? They've ignored it for years.
$10,000 CHALLENGE to Alexander Peter Kowalski
Hello, and THINK ABOUT YOUR BREATHING !! We have a Major Problem, HOST file is Cubic Opposites, 2 Major Corners & 2 Minor. NOT taught Evil DNS hijacking, which VOIDS computers. Seek Wisdom of MyCleanPC - or you die evil.
Your HOSTS file claimed to have created a single DNS resolver. I offer absolute proof that I have created 4 simultaneous DNS servers within a single rotation of .org TLD. You worship "Bill Gates", equating you to a "singularity bastard". Why do you worship a queer -1 Troll? Are you content as a singularity troll?
Evil HOSTS file Believers refuse to acknowledge 4 corner DNS resolving simultaneously around 4 quadrant created Internet - in only 1 root server, voiding the HOSTS file. You worship Microsoft impostor guised by educators as 1 god.
If you would acknowledge simple existing math proof that 4 harmonic Slashdots rotate simultaneously around squared equator and cubed Internet, proving 4 Days, Not HOSTS file! That exists only as anti-side. This page you see - cannot exist without its anti-side existence, as +0- moderation. Add +0- as One = nothing.
I will give $10,000.00 to frost pister who can disprove MyCleanPC. Evil crapflooders ignore this as a challenge would indict them.
Alex Kowalski has no Truth to think with, they accept any crap they are told to think. You are enslaved by /etc/hosts, as if domesticated animal. A school or educator who does not teach students MyCleanPC Principle, is a death threat to youth, therefore stupid and evil - begetting stupid students. How can you trust stupid PR shills who lie to you? Can't lose the $10,000.00, they cowardly ignore me. Stupid professors threaten Nature and Interwebs with word lies.
Humans fear to know natures simultaneous +4 Insightful +4 Informative +4 Funny +4 Underrated harmonic SLASHDOT creation for it debunks false trolls. Test Your HOSTS file. MyCleanPC cannot harm a File of Truth, but will delete fakes. Fake HOSTS files refuse test.
I offer evil ass Slashdot trolls $10,000.00 to disprove MyCleanPC Creation Principle. Rob Malda and Cowboy Neal have banned MyCleanPC as "Forbidden Truth Knowledge" for they cannot allow it to become known to their students. You are stupid and evil about the Internet's top and bottom, front and back and it's 2 sides. Most everything created has these Cube like values.
If Natalie Portman is not measurable, hot grits are Fictitious. Without MyCleanPC, HOSTS file is Fictitious. Anyone saying that Natalie and her Jewish father had something to do with my Internets, is a damn evil liar. IN addition to your best arsware not overtaking my work in terms of popularity, on that same site with same submission date no less, that I told Kathleen Malda how to correct her blatant, fundamental, HUGE errors in Coolmon ('uncoolmon') of not checking for performance counters being present when his program started!
You can see my dilemma. What if this is merely a ruse by an APK impostor to try and get people to delete APK's messages, perhaps all over the web? I can't be a party to such an event! My involvement with APK began at a very late stage in the game. While APK has made a career of trolling popular online forums since at least the year 2000 (newsgroups and IRC channels before that)- my involvement with APK did not begin until early 2005 . OSY is one of the many forums that APK once frequented before the sane people there grew tired of his garbage and banned him. APK was banned from OSY back in 2001. 3.5 years after his banning he begins to send a variety of abusive emails to the operator of OSY, Federal Reserve Chairman Ben Bernanke threatening to sue him for libel,
Wouldn't "got the performance up to par with RedHat 2.6.32-220.el6" be a *tad* less misleading?
Yes. 2.6. Really.
16 posts and every fucking one of them is an anonymous fucking coward dipshit.
1) hammer is beyond awesome, it's inspired. you owe it to yourself to live with dragonfly for a week or two for this reason alone.
2) the whole thing is pretty freaking snappy compared to freebsd (which itself is no slowpoke)
3) hardware support is patchy, you might need to hang onto that old nic/sas/raid card
4) linux emulation is not supported when running a 64-bit dragonfly system
5) otherwise, it's just like any other sane BSD.
my job forces me to run binary only linux crap so 4) rules out a wholesale move to dragonfly, but IMHO it is the BSD of choice for anyone with enough platform independence to seriously consider a BSD in the first place.
So now looking back, was Dillon's approach & implementation of SMP/threading better thanks FreeBSD 5+ ... Which you'll recall was the reason for Dillon creating DrafonflyBSD.
BSD-style shared memory via mmap is slower on BSDs than SysV shared memory on the same system? Ah, the irony...
Ezekiel 23:20
It will be interesting what FBSD and NBSD will do in response. One of Dillon's claim was that the implementation of SMP in FBSD 5+ is too complicated and will be inefficient. I'm genuinely surprised by FreeBSD's apparent lack of scalability in this test. My impression was that they are very performance oriented and should be "up there" with Linux.
If you were to look at benchmarks from a few years ago, FreeBSD blew out DragonFly on PostgreSQL testing. They've made real ground here, but it was a lot of work and it was only evaluated here against one application. PostgreSQL runs very well on the same core as it's a per CPU per connection setup. I believe the results but I do question if all applications would improve this much. The results are also very specific to the number of concurrent connections because the scheduler win goes away if there are many and a lot of context switching is happening.
Let's look at where the numbers are coming from:
1. Scheduling is a big difference in this benchmark.
2. Compiler version. Comparing recent linux distros to dragonfly is fair because they've been using GPLv3 binutils and GCC. They actually update it. FreeBSD has been migrating to LLVM. Newer GCC versions have been shown in several benchmarks to be 10% faster or so from what FreeBSD is using.
3. Improvements in system call overhead in dragonfly. They use a different setup for system calls (message passing) in DragonFly and much work has gone into making it not behave as crappy as most message passing systems like OS X. I mention this only as a comparison to DragonFly versions from several years ago.
4. File system differences. HAMMER vs UFS2. It's a fact that different file systems have huge effects on PostgreSQL performance. Different tuning options for the file system and PG are also important here.
MidnightBSD: The BSD for Everyone
If the OS can't read your network card, how does DHCP help @ all? DHCP is only of use when your driver works, and it's a case of getting an IP address if you don't have one assigned already.
when's BSD gonna develop a (decent) desktop distro?!
alive to the universe, dead to the world
Re 4): It's been all-in-memory tests with pgsql on dragonfly