Slashdot Mirror

You are a bastard, attempting to trivialize his contributions to push your cancerous social justice agenda. Out of all of the project contributors from all time, this guy is the fifth highest code committer. In a project with literally several thousand contributors, he's beyond the 99th percentile in contribution volume and many of his contributions were very significant to the LLVM project. But oh no, if he pushes some fucking special snowflake SJWs out by not caring that they aren't "FUCKIN' WHITE MALES" then "the downsides of having him on board outweigh the benefits." Ah well, safe spaces and "muh feelz" are far more important than lld having an ELF linker, right?

You need to go to your room while the adults talk about this, little one. How are you this stupid? And why is it always you, AmiMoJo, XXongo, and PopeRatzo that show up and spout the least rational SJW dumpster fires imaginable? You and your ilk's "views" consistently have zero basis in facts or rational or logical thought. You contribute nothing of value whatsoever and if you vanished from Slashdot there would be nothing of value that was lost.

Even without any sort of criminal record it's not a pleasant experience to enter the US, even as a Norwegian citizen entering via Canada. This May the robots routed me back form BSDCan (in Ottawa) through Washington, DC. It's possible that the fact that I did not apply for a visa (this was transit only, planning to stay on the ground roughly one hour between flights) complicated things a bit. As it turned out, in addition to the ordinary three forms (with more or less the same info in all of them) I needed to fill in a separate 'visa waiver form' (identical to at least two of the other forms in all other things than paper thickness, sheet size, color of paper and print and font) before getting to the fingerprinting, retina scanning and oral examination to check the validity of the information that I'd filled in, performed by a border guard who seemed to have been trained to appear hostile but was obviously monumentally bored by the whole process. This was after clearing the ordinary pre-boarding security theatre, mind you. And of course I would need to pick up the boarding passes for my connecting flights at the Washington, DC airport. That meant getting from one end of the airport to the other to pick up boarding passes and clearing another full act of security theatre in order to get back to where I could board the transatlantic flight. I did make my connecting fligh, running pretty much all the way except for the time spent lining up for the various security checks on the way. So yes, I can believe in a theory that US border control was a factor in deciding to place the next Olympics elsewhere.

I don't think that's a very fair characterization. According to PJD, who ported it to FreeBSD::

http://www.bsdcan.org/2007/schedule/events/43.en.html

See page 15 of the presentation. He calls it "highly portable". Perhaps it's the linux internals which are not modular enough!

I do like Quagga very much. But, its performance still doesn't quite match the Enterprise Cisco router. That said, Quagga works very well for small to medium sized businesses and Quagga may even outperform the lower end Cisco routers. The enterprise Cisco router has a slight advantage in that its hardware and architecture are designed for purely routing. I was bummed to find out that there was a performance gap. A Canadian University, University of Toronto, has a routing cluster based on Quagga. The administrator, Russell Sutherland, even said that UoT would be moving to a Cisco or Juniper router config as he said that he would need fewer Cisco units than Quagga servers to achieve the same amount of routing. The cost savings in power alone is not insignificant. It is a neat experiment and I hope that one day Quagga will surpass enterprise Cisco. Here is a PDF detailing what Russell Sutherland has done: Back to the Future: BSD on the Edge of the Enterprise.

I'm not an expert on the subject, but the parent post rather obviously isn't either, so I might as well answer some of the FUD there.

The reasons for the relatively rapid 5.x -> 6.x thing is well documented many places on the net. And the reason can be boiled down to what the parent said: 5.x had some performance problems, the jump to 6.x was intended to make clearer that a lot of those problems have been cleared up, even though 6.x is not really a huge change structurally from 5.x which preceded it.

A lot of this (and the ULE business) has to do with FreeBSD 5+ focus on SMP. DragonFly thinks it's being done in the wrong way. Fair enough. FreeBSD developers seem to think rather that it's too early to judge, as the work is not done. In 5.x the work was less done that some people thought, and this lead to many unfortunate things.

In 6.x its still not done, but it's a lot more done... performance is back, due to more key areas being freed from the "giant" kernel lock. Some interesting benchmarks were presented in the "ports monitoring" presentation by Mark Linimon at BSDCan. It seemed to show 6.1 approaching 4.x, after the performance dip of 5.x. But for SMP boxes 6.1 shines.

That being said, i have both 4.x and 5.x and 6.x boxes in production in different situations. I've had no stability problems with 5.x (except some ACPI issues on some boards), but I'd like to get them to 6.x if possible as soon as I can.

To abbreviate, as I am out of time: linux now more stable? Surely you jest. Perhaps it *can* be as stable, but I'm posting this message from my linux box (laptop) which has been tweaked and prodded, patched, and coaxed for years now and it's by far the least stable system I have. It's pretty stable at the moment. I have the magic combo of kernel patches it seems. But half the kernel revisions I install break something or other. Granted, it is a laptop, but still...

Anyhow, have fun on whatever O/S you prefer these days. If it works for you, that's great.

Saw Poul-Henning this weekend at BSDCan. Seemed just a friendly and approachable as he did when I was there two years ago. Besides the core development stuff he has some really interesting side projects as well which he gave presentations on: NanoBSD (a way to build and package BSD for use primarily on read-only embedded systems with flash config), and Varnish (a from-scratch new concept on for a high performance http cache focusing on CMS performance). Great stuff.

This idea is really cool, but implementing it by putting hooks into each device driver seems overly complicated.

Also, there's "GEOM gate" on FreeBSD: http://garage.freebsd.pl/GEOM_Gate.pdf
For other cool stuff with GEOM see here and here. See also this discussion thread about ggate's limits.

Without the paper to judge, without any testimony from any security authority, and without any obvious remedy, what would be gained by starting to generate kernel hacks at this time?

This is the same guy who calculated the 1 Quadrillionth hexadigit of Pi (no, not digit. It is in base 16). His project was called PiHex. According to his currently short but illustrious trackrecord, along with this current announcement, he is destined for being a big-name IT security guru.

Don't forget that BSDCan 2005 starts on Friday in Ottawa, Canada. Talks include:

The FreeBSD SMPng Network Stack -- Robert Watson
Live Network Backup -- der Mouse
"Because it has to be free" - Wireless support in OpenBSD -- Reyk Floeter
ioctl is soooo 1980ies -- Poul-Henning Kamp
A new security issue -- Colin Percival

Yes, you can still register.

Don't forget that BSDCan 2005 starts on Friday in Ottawa, Canada. Talks include:

The FreeBSD SMPng Network Stack -- Robert Watson
Live Network Backup -- der Mouse
"Because it has to be free" - Wireless support in OpenBSD -- Reyk Floeter
ioctl is soooo 1980ies -- Poul-Henning Kamp
A new security issue -- Colin Percival

Yes, you can still register.

http://www.bsdcan.org/

Basically, what they are doing is getting coffee shops to pay a small fee to host the access points (running a custom Linux configuration), networking those, and offering both the internet connection on the coffee house's dime and building out their own BBS-like intranet service.

Maybe the idea would be harder to get off the ground in other parts of the world, but if you can swing it, I think internet access is a big draw for people who otherwise may not bother.

BSDcan just needed a link....

OpenBSD really really supports it :) I went to a talk by Theo de Raadt at bsdcan this weekend, and about half the talk was about how they integrated this stuff into OpenBSD. Very informative, they're doing really interesting security work.

I'm surprised nobody mentioned that BSDCan started today at the University of Ottawa. If I wasn't working this weekend I would've gone (I think I'm only 3 or 4 hours away):(.