The current state of NetBSD

[sacerdos_daemonis]

Originally Posted by shep

No package builds in 5 months means either that no one tried or the builds failed and no one bothered submit patches.

I am not a programmer, so my question may be foolish, but does lack of new patches, or even lack of any activity, mean that a project is dead or obsolete? Or could it also mean there is nothing that needs to be fixed or changed? For example; how many changes, new code, etc. have been added to Openbox and Fluxbox over the last two or three years? I have no knowledge to comment on the specific projects being mentioned in this thread, but I, as a non-programmer, wonder how valid the argument is that lack of development activity translates to health of a project.

[Oko]

Originally Posted by sacerdos_daemonis

I am not a programmer, so my question may be foolish, but does lack of new patches, or even lack of any activity, mean that a project is dead or obsolete? Or could it also mean there is nothing that needs to be fixed or changed?

That is logically valid argument and might be even true for many times mentioned Amiga and Atari ports. However I am little scared to take your argument on the face value across the board. I will give you an example. I am somewhat familiar with NetBSD sparc64 port due to the fact that original OpenBSD support for sparc64 can be traced back to NetBSD. I still run a Blade 1500 in production and few years ago when I was selecting OS out of curiosity I compared the state of OpenBSD and NetBSD ports. It was like day and night. OpenBSD sparc64 had support for many more devices including the most advanced sparc64 processors produced by Sun Microsystem while NetBSD was stuck in late 90s.

Now lets talk about amd64 and ARM 64. Do NetBSD people really believe that those NetBSD ports are such a state of the art and that there are no new features to be added? I am really curios to hear examples from people who run NetBSD in real production (universities, labs, companies). What do you guys use NetBSD for and on what kind of hardware? How does network or file system(s) (for example NFS) performance compares to other BSD and Linux? How does Dom0 on NetBSD compares to the Debian one? Any cool appliances or specialized builds? How preform fully encrypted NetBSD installation? What is the current state of cryptographic drivers? What is the issue with software raid? Does NetBSD support soft RAID 6? How is the support for hardware raid? Does LSI MegaRAID Storage Manager run on NetBSD? Are there any issues with IPMI on NetBSD? What about net-snmp for example? What is the current state of Linux emulator in particular do they have 64 bit emulation? Can anybody run Linux MATLAB on NetBSD?

[ibara]

Originally Posted by shep

Stagnant code that is that is cycled through a build system is at best automated activity. Show me a commit to the NetBSD amiga port that is less than 5 months old.

Here are 22 files in the NetBSD/amiga port that, as of this writing, have commits to them younger than 5 months old:
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
http://cvsweb.netbsd.org/bsdweb.cgi/..._with_tag=MAIN
Please just stop. Continuing will just embarrass you further.

Originally Posted by shep

No package builds in 5 months means either that no one tried or the builds failed and no one bothered submit patches. The existence of 6.15 base amiga iso just means the build system churned one out. It does not mean it runs and it certainly does not mean that the NetBSD/amiga port is an active project.

Each reply makes it more and more obvious you don't know what you're talking about.
Surprise me, go do something for NetBSD.

[shep]

Here are the first two in your list

Revision 1.129 / (download) - annotate - [select for diffs], Sun Nov 16 16:01:40 2014 UTC (3 weeks, 5 days ago) by manu
Branch: MAIN
CVS Tags: nick-nhusb-base, nick-nhusb, HEAD
Changes since 1.128: +2 -4 lines
Diff to previous 1.128 (colored)
Remove unused extended attributes kernel options

As Masao Uebayashi pointed to me, UFS_EXTATTR_AUTOSTART, LFS_EXTATTR_AUTOSTART
and UFS_EXTATTR_AUTOCREATE are not used anywhere in the code. Remove them
as they have been obsolete for a long time:
UFS_EXTATTR_AUTOSTART was replaced by mount -o extattr
LFS_EXTATTR_AUTOSTART was created to match obsolete UFS_EXTATTR_AUTOSTART
UFS_EXTATTR_AUTOCREATE was replaced by sysctl vfs.ffs.extattr_autocreate

Revision 1.128 / (download) - annotate - [select for diffs], Wed Nov 12 10:47:21 2014 UTC (4 weeks, 2 days ago) by manu
Branch: MAIN
Changes since 1.127: +5 -2 lines
Diff to previous 1.127 (colored)

Support for UFS1 extended attributes in GENERIC and GENERIC-like kernels

This change just brings UFS1 extended attribute *support* in the kernel,
extended attributes are not enabled unless three conditions are met:
1) filesystem is UFS1 (newfs -O1)
2) .attribute/system and .attribute/user directories are created at fs root
3) filesystem is mounted with -o extattr

Some GENERIC kernels are obviously memory constrained, the extended
attributes options were not enabled for them, but just added commented out.
(kernel were considered memory constrained if QUOTA option was disabled)

Correct me if I'm wrong, but isn't this about the file system? Wouldn't each of the 52 architectures need these commits?

Each reply makes it more and more obvious you don't know what you're talking about.

You keep taking cheap, personal shots at me. I have not seen anything that tells me you are not beating a dead port.

[ibara]

Originally Posted by shep

I have not seen anything that tells me you are not beating a dead port.

Yeah you're done.

[fn8t]

Originally Posted by Oko

That is logically valid argument and might be even true for many times mentioned Amiga and Atari ports. However I am little scared to take your argument on the face value across the board. I will give you an example. I am somewhat familiar with NetBSD sparc64 port due to the fact that original OpenBSD support for sparc64 can be traced back to NetBSD. I still run a Blade 1500 in production and few years ago when I was selecting OS out of curiosity I compared the state of OpenBSD and NetBSD ports. It was like day and night. OpenBSD sparc64 had support for many more devices including the most advanced sparc64 processors produced by Sun Microsystem while NetBSD was stuck in late 90s.

Now lets talk about amd64 and ARM 64. Do NetBSD people really believe that those NetBSD ports are such a state of the art and that there are no new features to be added? I am really curios to hear examples from people who run NetBSD in real production (universities, labs, companies). What do you guys use NetBSD for and on what kind of hardware? How does network or file system(s) (for example NFS) performance compares to other BSD and Linux? How does Dom0 on NetBSD compares to the Debian one? Any cool appliances or specialized builds? How preform fully encrypted NetBSD installation? What is the current state of cryptographic drivers? What is the issue with software raid? Does NetBSD support soft RAID 6? How is the support for hardware raid? Does LSI MegaRAID Storage Manager run on NetBSD? Are there any issues with IPMI on NetBSD? What about net-snmp for example? What is the current state of Linux emulator in particular do they have 64 bit emulation? Can anybody run Linux MATLAB on NetBSD?

Well, if you get positive replies to much of that, I'll take one of my earlier statements back

I don't use Linux emu on NetBSD, but the Guide reports:

If you use a GENERIC kernel you don't need to do anything because Linux compatibility is already enabled.

If you use a customized kernel, check that the following options are enabled:

option COMPAT_LINUX
option EXEC_ELF32

or the following options if you are going to use 64-bit ELF binaries:

option COMPAT_LINUX
option EXEC_ELF64

when you have compiled a kernel with the previous options you can start installing the necessary software.

Off topic, I felt quite retarded that I didn't assume this:

There is another compatibility feature worth being aware of. NetBSD uses the same binary object file format across NetBSD machines with the same CPU. If you build binaries on your NetBSD sun3, you can run them on your NetBSD amiga, NetBSD mac68k, NetBSD mvme68k, NetBSD next68k...

Duh, right!

[ibara]

FWIW, the "building binaries on one platform to be used on all platforms of the same CPU arch" thing isn't unique to NetBSD. OpenBSD does this as well. I'd be hard-pressed to learn that others didn't do this as well.

[fn8t]

Originally Posted by ibara

FWIW, the "building binaries on one platform to be used on all platforms of the same CPU arch" thing isn't unique to NetBSD. OpenBSD does this as well. I'd be hard-pressed to learn that others didn't do this as well.

Right. No, I felt silly. Its like having the Keanu Reeves face and saying, "Whoa! It runs on the HP and the Dell". I get that way when I read late at night. After about two paragraphs I realize I forget every previous word, while reading the current one without any cognitive association to definition. Its probably good to confess these things once in a while so we're all on the same page about my competencies.

Wouldn't want any confusion about my confusion.

EDIT I blame it on Color Computer 2 & 3 exposure. That's right, I recorded that square bouncing ball to cassette.

[shep]

Seems like this thread has just been waiting to happen.

The following is a 2006 interview of Charles Hannum. Charles Hannum and Theo De raadt were two of the 4 NetBSD founders

O'Reilly 2006::Confessions of a Recovering NetBSD Zealot

Would you like to talk about the fork that originated OpenBSD?

Charles M. Hannum: No.

Since it came up in the /. thread, though, I would like to make one correction. It's widely claimed that I'm "the one" who ejected Theo from the NetBSD community. That is false. At that time in NetBSD's history, Chris G. Demetriou was playing the role of alpha male, and I wasn't even given a choice. I was certain it was going to bite us in the ass. I think the question for historians is not whether it did bite us in the ass, but how many times and how hard.

[ocicat]

Stop throwing rocks. It doesn't serve any larger purpose, & it is counter-productive.

It is also irresponsible.

Can blemishes in all projects be found? Of course. This isn't news, & dredging up past warts isn't newsworthy -- especially when those slinging the mud have no involvement in the project's direction. If you don't like how a project is being managed, do something constructive. Add to the discussion. This thread has added little value, if any at all.

Development is complicated, & the NetBSD project has done a masterful job of attempting to run on many platforms. Respect the problem, & if it seems to be a hinderance, learn about the real technical details that led to the decisions made. If learning takes time, so be it. You will be a better person for having done real homework, & an opinion will be formed on grounded reality.

Criticizing from afar with no real knowledge of the thorny issues involved is simply noise. Many of you have commented on this site for years, but what I see in this thread is truly embarrassing. This site has had a tradition of thoughtful discussion, but it is not demonstrated here.

[fn8t]

Originally Posted by shep

The following is a 2006 interview of Charles Hannum. Charles Hannum and Theo De raadt were two of the 4 NetBSD founders

O'Reilly 2006::Confessions of a Recovering NetBSD Zealot

Some of what Charles Hannum says is probably true. But, following his history in NetBSD and his involvement in the improper treatment of Theo, prior to the NetBSD/OpenBSD fork, he was also a part of the problem. Charles himself suggests Theo was treated badly, but his displeasure with NetBSD likely only partially agrees with the reasons Theo had sought for change within NetBSD. So here you have two different arguments against NetBSD's direction. When regarding the overall success or failure of the NetBSD project, both perspectives may declare NetBSD''s path toward an unsatisfactory result, but the dogmas for successful correction of that path don't necessarily agree. However, one could use every argument issued against NetBSD, regardless of its philosophy, to back up their own displeasure even if all the arguments themselves are not entirely reconcilable. Furthermore, one wouldn't have to actually have an argument to sustain. They could just stand on one side of the fence and throw anything they find usable to themselves from that side as ammunition. The person wouldn't even have to have had an initial reason for being on that side of the fence. They may have just landed there by dumb luck and played out the roll of their location. Often times this is just the initiation of growth, prior to mature opinion. Just about everyone starts out there, and hardly ever can anyone speak from any other location than this without restraint on their own expansion of education.

Edit I mean no disrespect to Charles Hannum

[Martillo]

I have an instance of NetBSD 7.0 BETA/99.2 running in VirtualBox (because I am not able to run it in bhyve).

[muflon]

Edit reason: concerned

[ibara]

I was somewhat hoping that shar wouldn't be found. Be extremely careful if you choose to read it.

[muflon]

Originally Posted by ibara

I was somewhat hoping that shar wouldn't be found. Be extremely careful if you choose to read it.

Obviously; 'it's hard to be a judge in your own case'.

[Carpetsmoker]

Late to the discussion, but I haven't been around...

In response to the "The Amiga port doesn't get many commits"-argument:

How many commits do you expect to see here? Amiga doesn't make new hardware (and hasn't for a while), so the platform-specific stuff probably supports all the desired features.
There are without a doubt bugs in the Amiga platform, but with little new hardware, most of the obvious one are solved already (I would be worried if there would a continues stream of bugfixes on such a platform).

In response to "NetBSD should drop old platform to speed up development"-argument:

Once you support more than 1 or 2 (very) different platforms, you need to write platform-independent code; adding a few extra platforms usually won't hurt you, neither do old platforms prevent the addition of new features, NetBSD has had loadable kernel modules for a while, and many features (like the new firewall, which was brought up as an example by someone) can simply be a loadable module; if a platform doesn't support that, then that sucks, but it doesn't prevent *other* platforms from using it.

Wikipedia mentions "Currently, unlike other kernels such as μClinux, the NetBSD kernel requires the presence of an MMU in any given target architecture.", NetBSD has also dropped some platforms, as can be seen on http://netbsd.org/ports/history.html
So it would seem that NetBSD doesn't go out of the way to support every possible platform out there.

[DaBSD]

Just my five pence to the discussion, also rewording some of what has been said above.. I think that NetBSD has suffered from the very fact that the BSD world is small after what's left of Linux and there's not enough community backing available for each version. Historians will tell us why FreeBSD became more popular than NetBSD around 2000, as for OpenBSD vs. NetBSD I think Open's proclaimed emphasis on security has been a much better advertisement than Net's record-breaking portability. You'd rather want your OS be secure than run on obscure hardware, wouldn't you.. And, as the BSD world is small and rather well interconnected, there's hardly been space for 3 or more equally strong OSes, with some pessimistic consequences for our dearly loved NetBSD.

P.S. One interesting comment from Facebook:

Pongthep Kulkrisada I use both NetBSD and FreeBSD. As far as I have tested, NetBSD is faster and cleaner. I always opt NetBSD.
5 December 2014 at 17:32 · Like · 1

Bartek Krawczyk Pongthep Kulkrisada, and what did you test and how?
5 December 2014 at 19:06 · Like

Pongthep Kulkrisada I do not adhere to any benchmark. I only test what I use. My family business is running local web server (intranet) using much CGI, which is mostly written in C. That time I tried both FreeBSD (9.x) and NetBSD (5.1.2). I could say NetBSD is 15-20% faster in all routines.
https://www.facebook.com/groups/2228...2987408348694/

[Oko]

Originally Posted by DaBSD

Pongthep Kulkrisada I use both NetBSD and FreeBSD. As far as I have tested, NetBSD is faster and cleaner. I always opt NetBSD.

What does it mean "cleaner"? Did he by any chance diff entire FreeBSD and NetBSD source three and have done code review of the difference? Speaking of which NetBSD is heavily relying on cross compiling and has almost non-existent QA.

[gpatrick]

Speaking of which NetBSD is heavily relying on cross compiling and has almost non-existent QA.

Where is your proof? Hard facts? Or is this more "OpenZFS and Solaris" are dying mumbo-jumbo?

[rocket357]

Good to see this thread isn't as dead as NetBSD is rumored to be. =)