Minix 3.2 Released, Uses LLVM/Clang, SMP, ELF

[kraftman]

This is similar to the problems with how modern Linux desktops are significantly less secure than Windows 7 and OS X (and even more so compared to Windows 8 and the next OS X release).

With Windows holes since a dos era it doesn't make it more secure.
And there's sandbox hole in OS X:

http://www.macworld.com/article/163591/2011/11/researchers_bypass_restrictions_of_os_x_default_sandbox_profiles.html

[kraftman]

With Windows holes since a dos era it doesn't make it more secure.
And there's sandbox hole in OS X:

http://www.macworld.com/article/1635..._profiles.html

@Elanthis

When comes to Linux I think AppArmor and SELinux is the answer to what you are asking, because they can also protect your home directory:

http://ubuntuforums.org/showthread.php?t=1008906

[ninez]

That was the point. In such case, where only my disk data matters, both systems are on pair when comes to 'unsafeness'. Micro kernel maybe has advantage when I care about uptime (but if there's cve in the kernel I'll have to put it down) or in the case RealNC described.

The example given by RealNC is contrary to what you say ~ because the data remains in memory, the file-system restarts, the data can be saved, and nothing is lost.. So they aren't on par, because if a driver crashes in linux, i get a kernel panic, and my data is lost.

For normal usage it's maybe not so noticeable (on mobiles....), but in the operations it will be much slower like the benchmark shows (so it's you who just assume something, because you didn't even show any benchmark and that's a fact there's more overhead in micro kernels design). Could you point me where the micro kernels are used when performance matters?

Actually, if you read back, i fully acknowledge that Microkernels have addition overhead. What i disagree with is that that overhead is 100X what you find in a monolithic kernel. ~ The best example of a Microkernel that is in high-use is QNX, unfortunately, to gain access to many benchmarks, most sites require membership (and i wouldn't be allowed to post them here anyway, so i don't get what you are wanting here). ~ I have however, looked at a few now. QNX 6.2 smoked Redhat's embedded linux in the 2000's (yrs). I also have an idea for you, why don't you go and look for benchmarks yourself, why don't you go to various companies websites who are using QNX Nuetrino ~ and see if any of their usages, apply to high-performance. Or maybe you could actually go and read a whitepaper or two, and count how many times you see the words 'high-performace' associated with QNX...

What I don't know exactly? If the bug is in the file system the risk is the same as with the monolitic kernels when comes to data safeness. Performance is also much worse with microkernels, so?

wrong, data safe-ness is supposed to be BETTER with Microkernels ~ that is what you are missing. And in some cases, performance loss is marginal, at best. (atleast with QNX this is the case).

[ninez]

LOL . NO i didn't assume anyone here knows about that pre BB purchase full free x86 developer RTP ISO unless you were around when Dan (qnx CEO) was involved with that AMIGA developer release when they nearly partnered commercially with his company.

Well, i'm glad i got a laugh out of you ~ that was exactly what i was trying to do. LOL.

perhaps i wasn't clear enough, im saying that that older 6.2 version was put together for full one time non commercial single x86 development platform given the AMIGA owners were considering both x86 and PPC at the time and qnx covers both those and ARM too, whereas their current BB limited developer ISO's are BB java centric and these newer versions are 30day trials (with an unstated little known fact that there is a free licence for NC end users) with a limited BB Java developer focus not generic x86 focus with GCC etc as the tool chain as back then, and not less featured as you put it for a commercial developer paying his fees to get that existing feature, but defiantly less featured as standard for the NC user at home, hence the advice to get that now while its still there and you dont need to register or have a personal NC key mailed to you as you do with the newer BB versions today.

I see. As far as i can tell QNX Nuetrino doesn't equal BBX ~ and what they are distributing on their website, is just straight Neutrino, no bells and whistles, just the base (companies would use) to build their system on. it's too bad that ISO you provided a link to, wasn't a live-cd or VM - otherwise, i might have ahd a look

[popper]

Well, i'm glad i got a laugh out of you ~ that was exactly what i was trying to do. LOL.

I see. As far as i can tell QNX Nuetrino doesn't equal BBX ~ and what they are distributing on their website, is just straight Neutrino, no bells and whistles, just the base (companies would use) to build their system on. it's too bad that ISO you provided a link to, wasn't a live-cd or VM - otherwise, i might have ahd a look

i dont usually read phoronix during the weekend so your lucky LOL, did you get and use the smaller one in that dir as well , thats the boot/install livecd, the one i mentioned shows the app's and source for them, I.E use the smaller one to boot, the bigger one for cd repository install to HD i think it was....

its been a long time since i vmware played/VirtualBoxed these but it does/did have a boot from cd and install option so you can make a VM or even install from windows if you ready wanted i seem to remember, and i also remember the user was probably root and a blank password to boot the livecd directly as that always got people confused back in the day.

[BhaKi]

I wonder if this guy ever regrets not releasing MINIX as open source before Linux?

MINIX existed before Linux but was closed source. Then came Linux which was open source and it became big and famous.
If MINIX had been open source from the start, then Linux would have never been written and Andrew S. Tanenbaum could be the rockstar that Linus Torvalds is today.

He does regret some things. But open-source/closed-source has nothing to do with it.

http://linuxfr.org/nodes/88229/comments/1291183

[kraftman]

The example given by RealNC is contrary to what you say ~ because the data remains in memory, the file-system restarts, the data can be saved, and nothing is lost.. So they aren't on par, because if a driver crashes in linux, i get a kernel panic, and my data is lost.

I was thinking about bug that will corrupt your data. It's maybe a corner case, but the point is there's always some risk. In Linux you can get kernel oops, but if this will safe your data I don't know.

Actually, if you read back, i fully acknowledge that Microkernels have addition overhead. What i disagree with is that that overhead is 100X what you find in a monolithic kernel.

In this case it's about process creation which was 140X slower in the benchmark. Other things were twice slower etc. so I'm not saying it's always hundreds times slower.

I also have an idea for you, why don't you go and look for benchmarks yourself, why don't you go to various companies websites who are using QNX Nuetrino ~ and see if any of their usages, apply to high-performance. Or maybe you could actually go and read a whitepaper or two, and count how many times you see the words 'high-performace' associated with QNX...

I'm just interested in HPC, enterprise, server and desktop usage and it's hard to find any QNX benchmarks. In others like RT systems it's maybe good.

wrong, data safe-ness is supposed to be BETTER with Microkernels ~ that is what you are missing. And in some cases, performance loss is marginal, at best. (atleast with QNX this is the case).

I know it's better overall when comes to your data safeness, but in the corner case I mentioned they're on pair.

[RealNC]

Let's keep in mind though that Linux is generally rock-stable. I may get a crash every couple of months, but consider that I'm running a bleeding edge installation (Gentoo, using latest testing (~arch) packages.) But even 5 crashes in a year (and that number is actually higher than what I really get) is acceptable for desktop use. And I don't think that the enterprise is running bleeding edge distros to begin with. On servers I administer, I run Debian stable. I can't remember when one of them last crashed. Actually I think they never crashed. Not a single time.

But... If I had a machine where crashes have bigger risks than just losing the download progress of your porn, like, I don't know, running a nuclear reactor or whatever, then I'd prefer a microkernel. But for desktops or even workstations? Nah. Linux is stable enough.

It's nice that Minix is there as an option, but it's doomed to obscurity on desktops. No one really needs it there. Even if it would overnight magically acquire all the features Linux or Windows has, I still wouldn't use it; I already get annoyed enough when I lose 2FPS in Skyrim.

[LinuxID10T]

Let's keep in mind though that Linux is generally rock-stable. I may get a crash every couple of months, but consider that I'm running a bleeding edge installation (Gentoo, using latest testing (~arch) packages.) But even 5 crashes in a year (and that number is actually higher than what I really get) is acceptable for desktop use. And I don't think that the enterprise is running bleeding edge distros to begin with. On servers I administer, I run Debian stable. I can't remember when one of them last crashed. Actually I think they never crashed. Not a single time.

But... If I had a machine where crashes have bigger risks than just losing the download progress of your porn, like, I don't know, running a nuclear reactor or whatever, then I'd prefer a microkernel. But for desktops or even workstations? Nah. Linux is stable enough.

It's nice that Minix is there as an option, but it's doomed to obscurity on desktops. No one really needs it there. Even if it would overnight magically acquire all the features Linux or Windows has, I still wouldn't use it; I already get annoyed enough when I lose 2FPS in Skyrim.

At work, we run Red Hat SANs. Up for 5 years would be more accurate than 5 crashes a year... Truth be told, in a production environment, stability is hardly an issue anymore. Hell, even Windows is stable enough for mission critical software. I'd say 99.9% of the time, if there is any downtime, it is due to either physical maintenance or software problems (software as in on top of the OS, not the OS itself.)

[rvalles]

Microkernels and overhead

I'd say to at least look at the "microkernel overhead talk" at:

http://fosdem.org/2012/schedule/track/microkernel_os_devroom