This wasn’t supposed to be leaked far and wide before c2k4:
(from the above link):
“As an aside, it irks me a bit to see this story posted before there was enough to report, especially since it was requested that this information be kept close to the breast until it was in a state for wider release. I even made a point of this in other thread here. on undeadly.
This isn’t to chew out phessler too much (he’s already got an earful elsewhere), but undeadly readers should be aware that this work is in preparation for something bigger heavy lifting exercises (for those who aren’t aware c2k4 is almost upon us). We have known about these efforts for a while, but had not posted them intentionally and I wish it hadn’t slipped out in this manner. If you want to see where efforts really started to become public on the mailing lists, check: http://marc.theaimsgroup.com/?l=openbsd-tech&m=108672757809273&w=2….
i appluad the SMP uptake in OBSD, but can we say Big Kernel Lock? its about 8 years too late really. Unless they make some HUGE strides, SMP will still be pretty useless in OBSD when you have giant locking the entire kernel down all the time.
zero scalability. lets hope they can get the locking pushed down really quickly to give us a good SMP branch.
This is why I have never really considerd OpenBSD as good for anything other than firewalls. Sure it could act as general server / desktop / workstation but FreeBSD is much better suited when security is not a #1. But anyway, glad to see that it has been done. I remember seeing something ages ago that two students at some university were working on this. Is this the fruit of their labour?
FreeBSD 3.x dosen’t have well SMP support. It probably won’t scale well on dual CPUs even.
The new NexGen SMP kernel for FreeBSD 5.x will eventually prove itself to be highly scalable and useful. However, it is still not finished. But rather than bitch about lack of SMP or whatever in open source operating systems, why not volunteer, its open source, you know
hee hee! very humerous! bit it is true. openbsd’s aims don’t really mix with being SMP. openbsd will be stronger and have a place if it sticks to being a secure and feature-less OS. yes, featureless. by that i mean cutting out features and APIs which are not necessary for use on secure servers. i think we can do without webcam drivers for OpenBSD (especially if they get cracked and used for spying!). even if reduced APIs mean some things don’t work. they need to be uncompromising on this. SMP is only pandering to the crowd. and the crowd are the great unwashed mediocrity. smp and applications are covered by other OSes.
Supposively, from what I’ve heard from FreeBSD advocates, FBSD 3.x/4.x scales better then Linux 2.2 did, but certainly not as good as 2.4 or 2.6. I have no idea how true that is, but for a dual CPU system its not all that bad.
What do you mean it’s a slow start? Of course it’s not going at the speed of light, but we all know that OpenBSD likes to verify each bit of code 30 times over, which is after all a good thing.
This wasn’t supposed to be leaked far and wide before c2k4:
(from the above link):
“As an aside, it irks me a bit to see this story posted before there was enough to report, especially since it was requested that this information be kept close to the breast until it was in a state for wider release. I even made a point of this in other thread here. on undeadly.
This isn’t to chew out phessler too much (he’s already got an earful elsewhere), but undeadly readers should be aware that this work is in preparation for something bigger heavy lifting exercises (for those who aren’t aware c2k4 is almost upon us). We have known about these efforts for a while, but had not posted them intentionally and I wish it hadn’t slipped out in this manner. If you want to see where efforts really started to become public on the mailing lists, check: http://marc.theaimsgroup.com/?l=openbsd-tech&m=108672757809273&w=2….
i appluad the SMP uptake in OBSD, but can we say Big Kernel Lock? its about 8 years too late really. Unless they make some HUGE strides, SMP will still be pretty useless in OBSD when you have giant locking the entire kernel down all the time.
zero scalability. lets hope they can get the locking pushed down really quickly to give us a good SMP branch.
they;ve got that NetBSD code to fall back on
This is why I have never really considerd OpenBSD as good for anything other than firewalls. Sure it could act as general server / desktop / workstation but FreeBSD is much better suited when security is not a #1. But anyway, glad to see that it has been done. I remember seeing something ages ago that two students at some university were working on this. Is this the fruit of their labour?
I remember seeing something ages ago that two students at some university were working on this. Is this the fruit of their labour?
No, they hired someone.
Indeed, I wonder how it compares to FreeBSD 3.x (sigh)
FreeBSD 3.x dosen’t have well SMP support. It probably won’t scale well on dual CPUs even.
The new NexGen SMP kernel for FreeBSD 5.x will eventually prove itself to be highly scalable and useful. However, it is still not finished. But rather than bitch about lack of SMP or whatever in open source operating systems, why not volunteer, its open source, you know
hee hee! very humerous! bit it is true. openbsd’s aims don’t really mix with being SMP. openbsd will be stronger and have a place if it sticks to being a secure and feature-less OS. yes, featureless. by that i mean cutting out features and APIs which are not necessary for use on secure servers. i think we can do without webcam drivers for OpenBSD (especially if they get cracked and used for spying!). even if reduced APIs mean some things don’t work. they need to be uncompromising on this. SMP is only pandering to the crowd. and the crowd are the great unwashed mediocrity. smp and applications are covered by other OSes.
SMP on OBSD has a long way to go, suffering quirks of applying new features to a kernel.
Let the bug hunting begin…
Supposively, from what I’ve heard from FreeBSD advocates, FBSD 3.x/4.x scales better then Linux 2.2 did, but certainly not as good as 2.4 or 2.6. I have no idea how true that is, but for a dual CPU system its not all that bad.
I’ve heard the exact opposite. As I understand it, FreeBSD 4.x and lower scale worse than Linux 2.2. And that 2.4 and 2.6 still scale much better.
FreeBSD 3.x has better SMP than Linux 2.0
Can someone who knows the facts for sure clarify?
I am quite happy that OpenBSD is finally getting SMP support. Even if it is a slow start.
What do you mean it’s a slow start? Of course it’s not going at the speed of light, but we all know that OpenBSD likes to verify each bit of code 30 times over, which is after all a good thing.
in my opinion, all FreeBSD prior to 5.x scales horribily on multiple CPUs.. I hear dragonfly work well though..I personally welcome SMP on openbsd
Ran for about 5 minutes, ran some DB work, and it crashed. D’oh! Trying again.