Where to find FreeBSD sub-projects?

I am new to BSD and exploring internals of the systems. I have seen OpenBSD system and its child projects and it shows more than half of the code or packages in the base system is its in-house softwares such as openssh, libressl, httpd etc. I am studying FreeBSD but cant find anything in handbooks or anywhere about in-house softwares or percent of code in the base system except bhyve and jails. As I said earlier, I am new to BSD so please ignore my misconception.
 
I know a few: NETGRAPH, VIMAGE, GEOM ( +GGATE/HASTD services ). But all this was born in FreeBSD 4/5 era and since then, it seems, nothing more was created. Except pkg, maybe - this is quite a recent development - from 2013y.
 
yes capsicum aswell still i am yet to find some new in-house packages or any innovations .
 
While the BSDs also share work between them (i.e FreeBSD has OpenBSD's pf and things like that), I would say that FreeBSD has been working on ZFS as an internal innovation.

Yes, I know ZFS came from Sun, but I think quite a bit of work has had to be done on it to make it work, or keep it working with Oracles latest version in Solaris.
 
While the BSDs also share work between them (i.e FreeBSD has OpenBSD's pf and things like that), I would say that FreeBSD has been working on ZFS as an internal innovation.
Yes, and it feels like FreeBSD is a lego, where every component from different sources are joined together to form a system which is customizeable according to the end user'
 
OpenBSD looks like a compact system as their team intends to make everything in-house so system works out-of-box in contrast to customization.
 
OpenBSD looks like a compact system as their team intends to make everything in-house so system works out-of-box in contrast to customization.

Be warned though, the OpenBSD project provides zero end user support and I mean really zero. If you have a problem with something on OpenBSD and you wish to make it known on their mailing lists you better have a patch ready that at least tries fix the problem, otherwise whatever you have will fall on deaf ears.

Their attitude may seem elitist and harsh but their position is that OpenBSD is an R&D project aimed at developing the various tecnologies associated with the project, a good example of that is OpenSSH.
 
Yes, and it feels like FreeBSD is a lego, where every component from different sources are joined together to form a system which is customizeable according to the end user'

Because a bunch technologies were ported from Solaris, Linux and Open/NetBSD? What is the ratio between ported code and native code in a FreeBSD base system? Aren't most userland components, 2/3 of drivers, kqueue, bhyve, ALTQ, the TCP stack, GEOM, the ULE scheduler, devd, vt (NewCons), CARP, DSBMD, gpart, the Linux API (now supporting musl binaries), UEFI boot, Lumina, package management tools (pkgng, portmaster, poudriere, synth), jail managment tools (ezjail, iocage), bhyve management tools (vm-bhyve, iohyve, grub2-bhyve), boot management utils (FreeBSD bootloader, beadm, bootparams, boot0cfg, bsdconfig), hardening utils (Capsicum, Geli, IPFW, SSP, nxstack, SegvGuard), networking devices (tap,lagg..), the sound system, and precious daemons (ftpd, dhclient, httpd, inetd, blacklistd, watchdogd, auditd, powerd, netif, routing, natd, moused, coretemp..) native of FreeBSD?
Since when is FreeBSD some sort of other operating system fork rather than a 386BSD/4.4BSD successor? Isn't rather OpenBSD a NetBSD fork? How much NetBSD code can still be found inside OpenBSD?
Is OpenBSD really different from FreeBSD in this to such an extent?
Weren't Solaris zones modelled after jails? oh right, jails someone forgot to mention them, the predecessors of all sort of operating system-level virtualization

Hasn't been FreeBSD one of the main OpenZFS contributors throughout all those years?
Porting technologies like DTrace, BSM, ZFS, Clang, ASan, sndio, PF, Wine, VirtualBox , NFS, GSS-API, Rust, AutoFS, Linux DRM graphic drivers.....requires effort, and time.
It always amazes me to see OpenBSD and Linux guys criticizing FreeBSD for having wonderfully imported ZFS from OpenSolaris eons ago, when Linux still relies on IRIX' XFS for production, and OpenBSD still has FFSv2 (originally FreeBSD's, McKusick's doing, like soft updates and background fsck) only, without support for snapshots and TRIM, unlike FreeBSD's.

And what about FreeBSD performance? Doesn't it require a lot of hard work to keep up and still be compatitive with systems like Linux and Windows, which thousands and thousands of more developers work on and many more companies finance?

More things like pkgbase, ZFS native encryption and RAID-Z expansion, bectl, encrypted boot, ASLR, PAX/mprotect, native SMB-2/3 support, Wayland, lubinput, Network Prefix Translation for IPv6 on IPFW, support for many aarch64 SoCs and RISC-V, as well as many drivers for improved hardware support are all likely coming with 12-RELEASE.

So tired of always having to scroll these topics on this forum, always telling the same old story, without solid evidence at hand, one just after the other. What do people expect from an operating system,isn't this enough?
 
DragonFlyBSD is where the most innovation seems to happen for now at the BSD camp
ln which ways concretely speaking? Outside the poorly portable HAMMER1/2 and their wonderfully scalable hybrid kernel what new techonologies did the Dragonfly bring to the BSD camp? IPFW3 perhaps?

[EDIT] you may also be interested on THIS list.
Have you ever seen some FreeBSD vs DragonflyBSD performance benchs on multiple cores, or are we just taking these words for granted? I remember having read some parallel comparing data and out of the 2 the 'winner' was all but clear, honestly FreeBSD seemed performing overall better.
IMHO one of biggest problem with FreeBSD project and community is not being able to show off and advertise themselves as other projects and their communities do. Reality is that though, while torrents of words are spent eulogizing those projects, FreeBSD runs for real on both small businesses and corporate environments
 
Just some: Swapcache, VKernel, HammerFS2 will be portable ( it is one of the goals ). Oh, their NFS implementation is unrivaled.

Reality is that though, while torrents of words are spent eulogizing those projects, FreeBSD runs for real on both small businesses and corporate environments

The subject is not about reliability but innovation, and at this point there is nothing innovative going on ( publically at least ).

Also, DragonFlyBSD is not supposed to be running on anything production because it is a research OS.
 
Just some: Swapcache
Didn't mention it as I always interpreted Swapcache just as an inevitable side effect of HAMMER and DBSD's SMP. Absolutely speaking, not very practical in my view, nor seriously convenient on other kernels and FSs.But I may be just rambling

if I got it right vkernel follows the 'anykernel' concept. How is this different from the portable (ported and tested on Slackware myself) lightweight and secure NetBSD's Rump Kernel? What advantages does it bring over it?
HammerFS2 will be portable ( it is one of the goals )
This is very good to know. I confess having heard this before already; however, words aside, since I have no means to analyze H2 code and establish whether or not it's portable (lacking professional C programming and *BSD kernel knowledge), I won't be completely satisfied until OpenBSD ports it as promised long ago.

Oh, their NFS implementation is unrivaled.
Better than Illumos', how? Or is just about *BSD?
 
The subject is not about reliability but innovation, and at this point there is nothing innovative going on ( publically at least )
But have been other BSDs more innovative in the last years? OpenBSD? Not from my point of view; otherwise NetBSD should be considered at least as innovative (e.g. Variexec, ptrace, KASan, MKPIE, SMAP, optimized SMP on aarch64). Outside unveil and KARL I fail to see how OpenBSD has been innovative lately. But all those aren't the kind of things I consider significant innovations capable of affecting global software development history, the way things like ZFS and jails did. Nothing like a DTrace, a SMF or a CrossBow has come out of those 2 meanwhile.

DragonflyBSD? OK, I'm in, H2 is already enough to consider it innovative. Unlike what my previous statements may suggest, I like DBSD and I've been a DBSD server user for some time, in spite of having only explored the surface its potential. H1 is where I focused my attention most.


Also, DragonFlyBSD is not supposed to be running on anything production because it is a research OS.
Which also accounts for the different development pace and the rather conservative approach FreeBSD may have to take

My whole point, anyway, is that outside DragonflyBSD, I don't see how other BSDs are or have been significantly more innovative than FreeBSD in the last decade. But even taking Dragonfly into account I fail in seing such a huge and allarming difference in terms of innovation. Is H2 actually different and better than APFS, Bcachefs and the other recently developed CoW FSs?

Side Note: OP mentioned OpenBSD's subprojects. How can LibreSSL, OpenNTPD, OpenSMTPD and the others be innovation? Is rewriting something from scratch in a arguably cleaner and more secure way being Innovative? Not to my eyes. The notable innovations OpenBSD brought throughtout its history are: PF, arc4random, strlcat, W^X, pledge.
 
Better read some docs: VKERNEL.

I am not finding the data about the DragonFlyBSD NFS implemention other than THIS, but it is utterly fast.

NFS V3 RPC Asynchronization - DragonFly sports a revamped NFSv3 implementation which gets rid of the nfsiod(8) threads and implements a fully asynchronous RPC mechanism using only two kernel threads. The new abstraction fixes numerous stalls in the I/O path related to misordered read-ahead requests.

-----

This is very good to know. I confess having heard this before already; however, words aside, since I have no means to analyze H2 code and establish whether or not it's portable (lacking professional C programming and *BSD kernel knowledge), I won't be completely satisfied until OpenBSD ports it as promised long ago.

Nobody would probably port it for now since it is still UNFINISHED.

My whole point, anyway, is that outside DragonflyBSD, I don't see how other BSDs are or have been significantly more innovative than FreeBSD in the last decade. But even taking Dragonfly into account I fail in seing such a huge and allarming difference in terms of innovation. Is H2 actually different and better than APFS, Bcachefs and the other recently developed CoW FSs?

IDK APFS, but Hammer and Bcache are very different. The later is primary focused on tiered storage.
 
Just for the record, autofs wasn’t ported - it was written from scratch, but in a way to make it similar from the user point of view to the OSX one.
 
Back
Top