FreeBSD Removes GCC From Default Base System

An anonymous reader writes "With the LLVM/Clang migration, FreeBSD developers have now disabled building GCC and the GNU C++ standard library (libstdc++) as part of the FreeBSD base system. GCC and libstdc++ have been superseded by LLVM's Clang and libc++, respectively, on primary architectures for FreeBSD 10.0." You can still flip a few switches to get GCC, but the system compiler will still be clang. Update: 09/11 14:50 GMT by U L : Reader Noryungi noted that the What's Cooking for FreeBSD 10 page is also worth a look, adding "I have to say, this is shaping up to be a very interesting release. Bhyve [the BSD hypervisor], in particular, sounds very promising."

Comments

Oblig, appropriate XKCD

[PSVMOrnot]

"the system compiler will still be clang"

Or the Onionequse version: FreeBSD's move to revolutionary new swording system adds a whole new meaning to compiling

Just one question

[Anonymous Coward]

WHY?

Re:Just one question

[Lendrick]

See this link for an explanation:

http://unix.stackexchange.com/questions/49906/why-is-freebsd-deprecating-gcc-in-favor-of-clang-llvm/49970#49970

In short, mostly it's due to FreeBSD's issues with the GPL, not all of which are purely philosophical (it affects their funding, for one thing). On the other hand, if you don't have a beef with the GPL, it's probably best to stick with GCC, which produces more performant code.

Re:Just one question

[kthreadd]

Especially GPLv3 which they don't want to include in their base OS. Ports are fine thouh since they are not part of the operating system itself. GPLv2 was never a problem.

Re:Just one question

[Anonymous Coward]

For many reasons.

Clang is faster
Clang produces faster code
Clang's license is BSD (and hence more in line with FreeBSD's philosophy)
Clang has built in static analysis tools
Clang's error messages are easier to understand than gcc's
Clang has better support for C++11 than gcc
Clang's code base is much less convoluted than gcc's, and easier to work on
Clang's code base is more modular, which allows you to easily use separate stages of the compiler in other tools
Probably a bunch of reasons I've forgotten too

The real question is actually, why are so many people stubbornly sticking to gcc when clang has surpassed it in pretty much every way.

Re:Just one question

[lactose99]

Licensing.

GCC in the FreeBSD base is stuck at v4.2.1 as that was the last version licensed under the GPLv2. As this is about 7 years old by GCC standards a newer compiler is a welcomed change and since CLANG is BSD-licensed it is more in line with the project's goals anyway.

Re:Just one question

[kthreadd]

Clang is faster
Clang produces faster code

Really?

Clang has better support for C++11 than gcc

They had almost equal support for C++11 when I check earlier this year.

Clang's code base is much less convoluted than gcc's, and easier to work on
Clang's code base is more modular, which allows you to easily use separate stages of the compiler in other tools

I think that depends on the reader.

The real question is actually, why are so many people stubbornly sticking to gcc when clang has surpassed it in pretty much every way.

Because it's a good compiler. Both gcc and clang are good compilers. They didn't stop developing gcc just because clang came around and started to make progress.

Re:Just one question

[Anonymous Coward]

Sorry, but it's you that looks like a douche.

Re:Just one question

[Anonymous Coward]

using less memory is also more performant, or needing less disk writes, etc.

Re:Just one question

[pipatron]

Execution speed is not necessarily the only metric used to judge a compiler. As a developer for embedded platforms I mostly look at code size, and you could also judge it on how well it handles various CPU architecture varieties.

Then again, your argument still stands. He could've written "better". ;)

Re:Just one question

[Anonymous Coward]

I know you are but what am I?

Re:Just one question

[Anonymous Coward]

I'd have to agree with the "anti-performant" sentiment. It isn't cromulent to say "performant" if don't want to sound like a douche. It makes it seem like you're trying to embiggen yourself.

Re:Just one question

[Anonymous Coward]

There are many performance-related measures of a compiler beyond speed like memory use or size of the binary.

Re:Just one question

[HornWumpus]

Yet another reason not to say 'performant', it's ambiguous.

Re:Just one question

[Anonymous Coward]

For once, that happens to be the perfect riposte.

Re:Just one question

[Anonymous Coward]

Are you paid by the letter? Never say/write 'more performant' it makes you look like a douche. Say 'faster'.

"more performancy"
"more performanceish"
"more performanceabilistic"

Re: Just one question

[Anonymous Coward]

Your MOM is a riposte

Re:Just one question

[Anonymous Coward]

It is not as vague or ambiguous as saying "better," so I guess we should have twice the vitriol for anyone who says something is better?

Re:Just one question

[1s44c]

Because GCC is a total mess internally and Clang/LLVM is neat, clean, maintainable, adjustable.

GCC is a dead end. LLVM is the future.

Re:Just one question

[Anonymous Coward]

Talking to yourself again, Anonymous Coward?

Re:Just one question

[1s44c]

The real question is actually, why are so many people stubbornly sticking to gcc when clang has surpassed it in pretty much every way.

I'm all for replacing GCC.
Does clang handle all the GCC extensions to C yet? Or is the (better) approach of getting rid of them being taken instead? The last I heard it was still a bit of a mess.

Re:Just one question

[HornWumpus]

It's exactly as vague and ambiguous as 'better', while sounding pseudo technical. Something a PHB would parrot.

Re: Just one question

[Anonymous Coward]

And a more performant one at that.

Re:Just one question

[Anonymous Coward]

I think that depends on the reader.

Not really actually. RMS actually even has admitted in emails that the lack of modularity, and rather spaghettish code in gcc is a deliberate attempt to make it hard for tools authors to integrate it, because he feared that a modular compiler would easily be used to make closed IDEs.

Re:Just one question

[Anonymous Coward]

WHY?

Why not?

Re:Just one question

[DuckDodgers]

I'm fuzzy on licensing. I know you can take FreeBSD and Clang/LLVM and make a proprietary product out of it. But can you fork them and release your fork under the GPLv3?

Maybe in all seriousness the best option for the Free Software Foundation and GNU Project in a few years would just be to make a GPLv3 fork of the most recent LLVM. The Phoronix benchmark put highly optimized GCC code as slightly faster than optimized LLVM code in a few cases, but in return for dramatically slower compile times. You're already working with C and C++, is a 4% performance improvement worth losing an extra few hours to week waiting for your compiler to finish?

Re:Just one question

[jellomizer]

Because RMS made it!!!

I think it is most likely due to the large set of OpenSource Apps written in GCC and people really don't want to risk trying a different compiler.

Re:Just one question

[Anonymous Coward]

Does clang handle all the GCC extensions to C yet?

Clang Extensions

Re:Just one question

[Anonymous Coward]

No that's just wrong. Better could also mean things like cheaper, less side-effects, doesn't kill puppies, etc. "Performant" limits the betterness to only performance factors.

Re:Just one question

[rubycodez]

GPL2 was also a problem, that was the current license when the various BSD started to explore alternate compilers. The BSD want the license for all wares in the core install to be BSD

Re:Just one question

[kthreadd]

I've not had that experience when reading the gcc source code. Maybe that used to be the case, gcc has been around for some time.

Re:Just one question

[ais523]

The AC used to be correct, but gcc have been trying to clean up more recently. (Most likely, at least partly because competition from clang meant that trying to lock out third-party closed-source plugins wasn't a particularly useful thing to do anyway.)

Re:Just one question

[slacka]

I'm deeply disappointed that this issue was decided over philosophical instead of technical merits. If Clang was superior to GCC in the majority of benchmarks, then I would support this decision. But that’s not the case, GCC is still leading in most benchmarks and can be an order of magnitude faster when the popular OpenMP library is used.[1] Sadly, BSD users are the losers here.

[1] http://www.phoronix.com/scan.php?page=article&item=llvm_33svn_competes&num=5

Re:Just one question

[Chemisor]

My reason not to use clang: clang produces larger code than gcc. 10-20% larger, in many cases. For some reason gcc developers are much better at size optimization.

Re:Just one question

[anasciiman]

Faster isn't the same as higher performance.

Re:Just one question

[Ash Vince]

I'm deeply disappointed that this issue was decided over philosophical instead of technical merits. If Clang was superior to GCC in the majority of benchmarks, then I would support this decision. But that’s not the case, GCC is still leading in most benchmarks and can be an order of magnitude faster when the popular OpenMP library is used.[1] Sadly, BSD users are the losers here.

[1] http://www.phoronix.com/scan.php?page=article&item=llvm_33svn_competes&num=5

So what would you prefer? That BSD threw out their entire licensing policy and adopted the GPL3? Or that they stuck with an old version of GCC that was licensed under the just about permissible to them GPL2 but that got no updates.

Maybe they could have forked GCC then tried to maintain a version that was licensed under GPL2 and backported GPL3 GCC fixes into their fork but that strikes me as being a legal nightmare to be honest.

It would have been more useful it phoronix compared LLVM/Clang 3.3 to GCC 4.2.1 and you posted that as that is the only thing the BSD community could use instead.

This decision was forced by the GCC adopting the GPL3 but was probably always going to happen sooner or later. The BSD community and the FSF have VERY different ideas of what constitutes "free" software with the FSF actually wanting their code to be less free in order to make sure that if you use it you have to let other people see how you use it and keep any derivatives free.

Re:Just one question

[kthreadd]

True. I should have written that it wasn't a problem to include it in the distribution, even if it wasn't optimal.

Re:Just one question

[gnupun]

Not really actually. RMS actually even has admitted in emails that the lack of modularity, and rather spaghettish code in gcc is a deliberate attempt to make it hard for tools authors to integrate it, because he feared that a modular compiler would easily be used to make closed IDEs.

Isn't hard to read code more or less the same as closed source code?

Re:Just one question

[lgw]

OK, really: do you seriously expect BSD not to stick with BSD Licensed code? It's pretty much the point of the exercise. YouBetterSayGNULinux is there if you want it - you can have all the GNUs you want. BSD is not and should not be a Linux distro - it's its own thing, older and with a different mindset.

Re:Just one question

[HornWumpus]

Cheaper is a performance metric with money as the denominator. The others are reaches that would be excluded by context.

Re:Just one question

[10101001 10101001]

... and since CLANG is BSD-licensed it is more in line with the project's goals anyway.

"The goal of the FreeBSD Project is to provide a stable and fast general purpose operating system that may be used for any purpose without strings attached."

So, yes, more in line with the project's goals. But I'm still waiting for FreeBSD to switch to WTFPL, since clearly the whole disclaimer of warranty is a very-much strings attached component of BSD licensing that I don't see ever going away. After all, given that a big part of FreeBSD support is corporate funding and corporations just love suing each other over all sorts of ridiculous things, included things like implied warranty.... Really, the BSD licensing is just tying their hands.

Re:Just one question

[hedwards]

You can't relicense somebody else's code. You can use the code however you like, but you're required to use the code only with the license intact.

It gets a little bit odd when you mix GPL code into the mix. But, the GPL code is usually segregated to keep it from contaminating the rest of the code. So, when you compile it the resultant binaries would be GPL, but only the portion of the source code that was GPL is GPL. Or at least that's how it was under the GPLv2, not sure what GPLv3 really does there.

Using APIs and ABIs that are based upon GPL code does not require that the binaries be GPLed either. I'm not quite sure as to what the problem here is with the GPL, but it's a GPLv3 specific problem.

Re:Just one question

[Anonymous Coward]

Your linked comparison isn't relevant to the FreeBSD situation. The GCC version in FreeBSD is 4.2.1, the last available under GPLv2. Clang is overall an improvement In comparison to that version.

http://www.phoronix.com/scan.php?page=article&item=freebsd_91_llvmgcc&num=2

Re:Just one question

[BitZtream]

Clang is faster
Clang produces faster code

Really?

Yes, well known and accepted. The craptastic GCC developers have been hurting its improvements for years. They have serious control freak issues and as such, GCC has fallen well behind the curve. This isn't a new topic of discussion.

Clang has better support for C++11 than gcc

They had almost equal support for C++11 when I check earlier this year.

Almost ... is not the same, and which parts are important.

Clang's code base is much less convoluted than gcc's, and easier to work on
Clang's code base is more modular, which allows you to easily use separate stages of the compiler in other tools

I think that depends on the reader.

Only if you're a fanboy. GCC is well known to be an ugly ass code base for multiple reasons. This is an OLD discussion.

The real question is actually, why are so many people stubbornly sticking to gcc when clang has surpassed it in pretty much every way.

Because it's a good compiler. Both gcc and clang are good compilers. They didn't stop developing gcc just because clang came around and started to make progress.

GCC hasn't been a 'good' compiler unless you're only metric is 'its GPL'. clang still has a long way to go though, both are shit compared to commercial compilers, many of which make gcc and clang look absolutely pathetic in almost every way. Hell, the MS compiler is light years more advanced than GCC when it comes to the code it produces.

You're just a fanboy.

Re:Just one question

[Cramer]

They're better at optimization - period. The only people that do it better are the people who built the processor. (read: intel -- icc, but don't try building the linux kernel with it, as there's way too much "gcc-ness" in the source code.)

(Also, Sun made a better sparc compiler than gcc, back in the day)

Re: Just one question

[Anonymous Coward]

In general, the reason for the switch was the openly agressive and rude treatment from the GNU people went it came to accepting changes or adopting requirements from other camps.
Clang is performing better than gcc 4.2.1, the last compiler from the FSF with a license normal people can understand. It is also has much better diagnostics and c++ support so it is a positive change.

Re:Just one question

[dnaumov]

Technically not true. There is lots of BSD-originated code in proprietary systems. The Windows version of the command-line FTP client that ships with Windows is a good example.

Re:Just one question

[lister king of smeg]

I'm deeply disappointed that this issue was decided over philosophical instead of technical merits. If Clang was superior to GCC in the majority of benchmarks, then I would support this decision. But that’s not the case, GCC is still leading in most benchmarks and can be an order of magnitude faster when the popular OpenMP library is used.[1] Sadly, BSD users are the losers here.

[1] http://www.phoronix.com/scan.php?page=article&item=llvm_33svn_competes&num=5

So what would you prefer? That BSD threw out their entire licensing policy and adopted the GPL3? Or that they stuck with an old version of GCC that was licensed under the just about permissible to them GPL2 but that got no updates.

Maybe they could have forked GCC then tried to maintain a version that was licensed under GPL2 and backported GPL3 GCC fixes into their fork but that strikes me as being a legal nightmare to be honest.

It would have been more useful it phoronix compared LLVM/Clang 3.3 to GCC 4.2.1 and you posted that as that is the only thing the BSD community could use instead.

This decision was forced by the GCC adopting the GPL3 but was probably always going to happen sooner or later. The BSD community and the FSF have VERY different ideas of what constitutes "free" software with the FSF actually wanting their code to be less free in order to make sure that if you use it you have to let other people see how you use it and keep any derivatives free.

gcc being gpl v3 does not effect the output of the complier only code linked to or borrowed from it

Re: Just one question

[dkf]

[Clang] is also has much better diagnostics and c++ support so it is a positive change.

It's also a lot more modular, reputedly (I've not verified that myself), and seems pretty fast to build. Mind you, I'd only really be interested if they were to work more on the stability of the ABIs of critical support libraries. Having to recompile for each version of libstdc++ was really annoying, given that maintaining a whole Linux distribution was never a desired personal goal of mine. Being able to ship the same binary to all users is nice, but even just one version per broad platform is better than what happens all too often.

I think it's a shame that open compiler projects can't share more, even if just their techniques and not the code itself. I can see that the difference in license and philosophies behind them would push the camps apart though.

Re:Just one question

[fast turtle]

Several reasons:

1) GCC dropping support for so called Obsolete Hardware that Freebsd still supports. This makes it impossible to use the same compiler version across all hardware.

2) GCC optimizations do not 100 percent produce identical binaries due to optimizations chosen by the compiler. This is a critical Security Issue that has been addressed with LLVM/Clang - speed isn't important at this time. A toolchain that produces 100 percent duplicate binaries was the requirement on all hardware that BSD supports.

3) GPL issues - don't know/don't care as the Second issue is critical while the first was very important. Some of the hardware that BSD still supports isn't supported by anything later then GCC 2 (that's 15-20 year old hardware) think it's some mission critical systems they're being paid to support so worth the time/effort to solve the hardware support removal issue by the GCC devs.

Guess what, you can't even use i386 for GCC anymore because they're supposedly obsolete yet they're still mission critical for the U.S. Government and Miltiary systems, meaning it is not obsolete but the GCC team removed support for the hardware anyway. Well since the Government isn't paying for support to GCC, Guess we can remove hardware that's a pain to support and isn't "Oh Shiny!" That was one of the big reasons given for the LLVM/Clang project 5-7 years ago AfaiR - found it interesting at the time and hoped they'd succeed as it apears they did.

Re:Just one question

[Darinbob]

I sort of have to stick with it, since I need it as a cross compiler, LLVM is not very tried and true with non-Intel processors. Plus it's a lot of code to patch up to make it work. Inline assembler and stuff like that. The "gcc" on MacOS was completely bonkers since it claims with preprocessor macros that it's GCC and yet it uses LLVM on the backend which broke code (again, inline assembler).

Utlimately so many systems rely on GCC because it's been the de-facto standard for so long. Build scripts that parse for errors and warnings, "well known" command line options, and so forth. Be interesting to see how much stuff breaks.

Re:Just one question

[hedwards]

Sure it's true. And that's why for years MS Windows had that BSD advertisement clause included in the software.

Re:Just one question

[evilviper]

I'm deeply disappointed that this issue was decided over philosophical instead of technical merits.

So you think we should scrap GCC all-together as well, and all switch over to Intel's C Compiler (ICC), since it's technically better?

After all, you think philosophy doesn't matter...

And GPLv3 incompatibilities with the project are more than just abstract "philosophical" differences, as it can impact use and distribution of FreeBSD.

Re:Just one question

[Kohlrabi82]

Useful error messages are a technical merit. It makes development much easier. So they traded run-time vs. development time here.

Re:Just one question

[serviscope_minor]

Clang produces faster code

Except according to all benchmarks.

Clang's error messages are easier to understand than gcc's

If you compare current Clang to old GCC then yes. If you use a new GCC, you'll see that it has much improved the error messsages.

Clang has better support for C++11 than gcc

Well known fact: clang supports 110% of C++11 compared to GCC's 100%.

The real question is actually, why are so many people stubbornly sticking to gcc when clang has surpassed it in pretty much every way.

Because GCC has a more complete frontend, better extensions, a better optimizer and much better OpenMP support.

Re:Just one question

[Dog-Cow]

Most OS X and iOS programmers have been using llvm/clang for the past year or so, on both x86 and ARM. Xcode 5 (GM seed available now) only supports clang. GCC is officially out for Apple development.

Re:Just one question

[Darinbob]

Yes, but you can use Apple computers for stuff other than Apple development. I am not a MacOS or iOS programmer, except as far as creating in-house tools.

The real problem though is the weird hybrid that Apple added. /usr/bin/gcc is not GCC and is also not LLVM, but a mix of the two. If you use "gcc" it claims to be GCC-4.2.1 even though it rejects some code that should work for gcc 4.2.1. I'd have been happier if they just got rid of gcc altogether rather than leave around a legacy version that doesn't work as it caused a lot of wasted time and confusion.

Re:Just one question

[smash]

Benchmarks are nothing to do with it. CLANG is more modular, generates better compiler warnings, plays nicely with IDEs and works better with Objective-C, for a start.

Benchmark performance can be improved with time. CLANG was started mostly because the GCC maintainer(s) were not interested in improving the objective-c support.

Re:Just one question

[smash]

Have you read the CLANG source to compare?

Re:Just one question

[armanox]

Sun's compilers are still available (last I checked) for Solaris on SPARC, Solaris on Intel, and Linux on Intel.

Re:Just one question

[willy_me]

The real problem though is the weird hybrid that Apple added. /usr/bin/gcc is not GCC and is also not LLVM, but a mix of the two. If you use "gcc" it claims to be GCC-4.2.1 even though it rejects some code that should work for gcc 4.2.1. I'd have been happier if they just got rid of gcc altogether rather than leave around a legacy version that doesn't work as it caused a lot of wasted time and confusion.

Standard GCC 4.0, GCC 4.2, and 4.2 with LLVM backend are installed with XCode - at least the current version. Calling "gcc" executes the version with the LLVM backend. If you want the standard GCC backend, just call "gcc-4.2" in place of "gcc".

Complaining that Apple has provided too many compiler options is a little ridiculous. Developers should be able to figure these things out. One expects Apple to simplify things for regular users; but as a developer, I appreciate that various options were made available.

Re:Just one question

[Anonymous Coward]

I'm a FreeBSD user and a C programmer and you know what? I haven't lost anything.
Clang has been my main compiler for several years, it's just much more pleasurable to work with. Performance? When doing performance testing, I always use several different compilers because I don't want to overtune for any one. The latest gcc comes on top more often than clang does. But the gcc in base (4.2) is a sucker, clang beats it usually.
Oh, BTW, none of programs that are important performance-wise for me uses OpenMP.

Re:Just one question

[Anonymous Coward]

I'm fairly certain that there is no OpenMP code in the kernel or the base utilities, and unlikely to be for the foreseeable future. If you want contemporary OpenMP compilation then current gcc is in ports for you. See: everyone wins.

Re:Just one question

[Anonymous Coward]

Re: GCC extensions: many of them but not all. Clang has stated (see the (fairly poor, IMO) online manuals) that it will never support gcc's "register globals" for example, which probably makes some of the fast threaded forth implementations out there non-starters. Most of the __builtin_foo() functions are there though, and there are some lovely extensions of its own, including a nice abstraction of vector operations, which makes it particularly suitable to cross-platform SIMD and OpenCL optimisations.

Re:Just one question

[Ash Vince]

gcc being gpl v3 does not effect the output of the complier only code linked to or borrowed from it

So? This is about them including the compiler in the base system. Or are you suggesting they include a completely different complier as part of the base install to the one they actually use internally to build the system?

Re: Just one question

[Anonymous Coward]

Your mom is performing ant? Performing ant... Performing aunt? So... Your mom is her own sister and she's a stripper, right?

Re:Just one question

[Anonymous Coward]

You're a moron. GCC is not the Linux kernel.

Sure, the Linux kernel dropped support for i386, because you're a moron if you're trying to run modern applications on 30-year-old hardware. GCC still compiles code for 386 just fine.

Re:Just one question

[dnaumov]

Which sure as hell doesn't mean ftp.exe is licensed under the BSD license.

Re:Just one question

[Bengie]

BSD gets imported into Linux quite often. The only requirements for BSD 2 clause are these

Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

Since GPL already requires the source distributed and the BSD clause just requires the copyright notice, it plays nicely.

Re:Just one question

[Anonymous Coward]

Neither GNU or the GPL have anything to do with Linux in this regard.

Re:Just one question

[dmoen]

That benchmark is old. OpenMP support for Clang was announced last week.
http://openmp.org/wp/2013/09/clang/

OpenBSD 10 hasn't been released yet, so it is premature to say that "BSD users are the loser here".

Re:Just one question

[TheRaven64]

Performance isn't the only metric by which you judge compilers, although I agree that it's an important one. OpenMP support affects a very small number of programs. Intel has just open sourced (BSDL) their OpenMP runtime and is pushing patches into clang, so that support should appear soon. The rest of the benchmarks show very little performance difference between clang 3.3 (which is in FreeBSD 10.0) and gcc 4.8, and neither is a clear winner in all cases. In comparison to the last GPLv2 GCC (4.2.1), which FreeBSD currently ships, both are significantly faster. If you need OpenMP today, gcc 4.8 is in ports, so just install the lang/gcc48 port and use it.

However, as I said, performance is not the only metric. Clang has full support for C++11 and supports more of C11 than GCC (for example, it correctly supports _Atomic and it supports _Generic). In FreeBSD, we're now using _Generic in a number of system headers and so you get better error reporting if your compiler supports them (fallback code exists if it doesn't). We found a few logic bugs in ports as a result of this, some of which have existed on other platforms and compiled without warnings for years. So, if you want to use the latest versions of C, C++, or Objective-C, clang is a better choice for you. We're also now shipping libc++ as the default C++ standard library, so we have a complete C++11 stack.

If you're developing C-family code and don't have a Coverity license, then having the clang static analyser around is also very helpful (it's built into clang, so you get it as part of the default install).

Re:Just one question

[TheRaven64]

I was a clang committer before I joined the FreeBSD project. I became a clang committer because I wanted an Objective-C compiler that supported a modern dialect of the language for non-Apple platforms. I took a look at the Objective-C code in gcc. It mingles parsing, semantic analysis and code generation in a convoluted knot. After a week of failing to understand it, I took at look at clang. At the time, Clang has parsing support for Objective-C, but no code generation support at all. Within a week, having never looked at the Clang codebase before, I had it generating code for a superset of the language that GCC supported.

I've also use clang to do syntax highlighting, to inspect files for type information for use in an FFI for a JIT-compiled language, and as a static analyser. How many tools have you seen that are based on GCC and aren't compilers?

Re:Just one question

[TheRaven64]

Yes, well known and accepted. The craptastic GCC developers have been hurting its improvements for years. They have serious control freak issues and as such, GCC has fallen well behind the curve. This isn't a new topic of discussion.

I'm a clang developer and you're a troll. Clang is faster than GCC on some benchmarks. GCC is faster than clang on others. The new autovectorisation code in Clang is usually faster than GCC, sometimes a lot faster, occasionally slower. Clang doesn't yet have the OpenMP support upstream. In most real-world code you won't see a difference of more than 10% between them, and even then you won't see a consistent winner.

I don't like GCC, and I don't use GCC (except on MIPS, where clang isn't quite ready), but GCC is not 'well behind the curve', it still gives acceptable performance. It was never the fastest player (ICC, XLC and Open64 were all faster), but it's pretty respectable.

Re:Just one question

[TheRaven64]

And (at least legally, there may be some exceptions), none of the code in these proprietary products was relicensed. It is built and shipped in accordance with the terms of the license, which permits redistribution in proprietary form as long as you provide attribution, don't sue the original developers if it doesn't work, and don't use their names to advertise your product. The fact that the overall work may have a different license from its components is a different issue. This is why detractors call the GPL 'viral', in that it (more or less) requires any work that contains GPL'd components to be GPL'd.

can it build the linux kernel?

[Anonymous Coward]

Can LLVM's Clang build the Linux kernel and QT/KDE?

I would love to have a Linux system,(not android), completely free of GNU.

Re:can it build the linux kernel?

[Lendrick]

Good luck with that. If you replace the (GPL'ed) Linux kernel, then it's not Linux anymore.

Re:can it build the linux kernel?

[kthreadd]

The Linux kernel itself is not GNU. Removing all GNU software will not remove the kernel.

Re:can it build the linux kernel?

[Chrisq]

I would love to have a Linux system,(not android), completely free of GNU.

That's a bit like wanting a Windows 8 system completely free of Microsoft licensing ... not going to happen

Re:can it build the linux kernel?

[dosius]

You'd want another libc too, like MUSL.

Re:can it build the linux kernel?

[kthreadd]

Can LLVM's Clang build the Linux kernel and QT/KDE?

They are getting close to building the Linux kernel according to these slides. Qt has been able to use Clang for a long time now, but I don't know about the entire KDE.

Re:can it build the linux kernel?

[Anonymous Coward]

He said GNU, not GPL.

Re:can it build the linux kernel?

[Anonymous Coward]

That's why it's called GNU/Linux, not just Linux.

Re:can it build the linux kernel?

[andreicristianpetcu]

Why? The GPL code is the biggest technological common pool ever. I love the GNU project! The weak copyleft licenses are not a common pool. They are scattered projects from a legal point of view. Look what happened to khtml and kjs they helped build Safari which is proprietary.

Re:can it build the linux kernel?

[wagnerrp]

GNU the operating system, not GNU the public license...

Re:can it build the linux kernel?

[andreicristianpetcu]

Also, GPL code will always be ahead of weak copyleft code. Take Libre Office vs Open Office. Libre Office can take code from Open Office but OO can't take code form LO. Code that LO takes from OO will always be free.

Re:can it build the linux kernel?

[Anonymous Coward]

Why on earth would you want a "common pool"? If anything, I'd be more likely to avoid such a thing.

And yes, it's terrible what happened to khtml, it got turned into the world's most popular open HTML engine. Just think, if it had been GPL, it could have remained obscure and unused forever!

Re:can it build the linux kernel?

[jedidiah]

You are trying to flee the spirit of the FSF in Linux.

It's much like trying to avoid the lingering legacy of DOS in Windows.

If you are a wannabe software Robber Baron, then Linux just isn't for you. Finally after all this time the anti-GPL crowd have their own compiler and they're using it. No big surprise there.

Re:can it build the linux kernel?

[Anonymous Coward]

No, the Linux kernel is not called GNU/Linux.

Re:can it build the linux kernel?

[andreicristianpetcu]

They broke it of from the community. They made changes but they did not commit the diffs. This after huge pressure from KDE devs. I prefere to have my code eaten up by other GPL projects and remain free forever than get assimilated in Apple's pretty iShit.

Re: can it build the linux kernel?

[Anonymous Coward]

That iShit turned into Chrome and as far as I know both WebKit and Blink is free software.

Re:can it build the linux kernel?

[NatasRevol]

Not exactly Safari, but pretty much what khtml became. And it's still open source. From Apple.

http://opensource.apple.com/source/WebKit/WebKit-7536.25/

Re:can it build the linux kernel?

[Anonymous Coward]

Wait a minute... if you say "Linux kernel" that's like saying "kernel kernel". You should have said "No, Linux is not called GNU/Linux" or, more accurately, "No, Linux in GNU/Linux is not called GNU/Linux."

Re:can it build the linux kernel?

[tlhIngan]

Also, GPL code will always be ahead of weak copyleft code. Take Libre Office vs Open Office. Libre Office can take code from Open Office but OO can't take code form LO. Code that LO takes from OO will always be free.

And that is one of the worst justifications ever. In fact, it's so bad, it lends credence to the "GPL is viral" belief - because the GPL can take and take and take, and not give back (which one could argue is against the SPIRIT of FOSS - that one shares so everyone benefits). Yes, it's legal, just TiVoization and GPLv2 is legal, but spiritually, it really reeks, especially when GPL advocates claim moral superiority.

Heck, perhaps the biggest problem with BSD is NOT the ability for "software robbers" to take and close the source, but for "software robbers" to take and claim moral superiority that their license "frees" the code (funny how "freeing" or "liberating" the source involves ensuring that it can never get back to the original authors, eh?).

Personally, I use unmodified BSD or GPLv2, because I have objections to the GPLv3 (and no, you can't mix v2-only and v3 code because the GPLv3 can be seen as causing restrictions disallowed by GPLv2. Funny how the champions of Freedom even admit that there are limits to how Free you can be).

Re:can it build the linux kernel?

[Anonymous Coward]

Why are you so rude to RMS? What have you done to make the world a better place? Now get off my lawn you little twit.

Re: can it build the linux kernel?

[andreicristianpetcu]

Chromium is free software. Google Chrome and Safari are not. If Google did not want to publish Chromium, it did not. The LGPL license is weak copyleft not strong copyleft.

Re:can it build the linux kernel?

[andreicristianpetcu]

Apple is the bully who took your 3 lunch sandwiches, added ketchup to one of them and gave it back to you and made you say "thank you sir for the ketchup".

Re:can it build the linux kernel?

[santax]

Which I and probably about every other BSD-style coder respects. However, this debate is really old. I think there is something to be said for both license-models and it is good they both exist. I however do feel that true free code, is well... really free. Also to be used without giving the source. It's a choice I made and I can perfectly understand why people don't want to have Big Ol' Apple stealing their stuff and making heaps of money from it. But, I can also understand why people want to really give the code away. It's all about perspective and there is no right or wrong here. It all depends on the personal views of the coder/owner of the code and what that person/org wants to let others do with their code. BSD has the advantage that it can become closed source. That makes it easier to generate profit, but also (especially in the commercial world) there are other good reasons why people want to protect their product. GPL has the advantage that every time someone uses your code, that code will become available also. Just don't ever think that because your code is GPL'd, no one will dare to use it in close source. As much as I would love such a world, that would be very naive.

Re:can it build the linux kernel?

[andreicristianpetcu]

BSD think of the freedom of the distributor, the GPL thinks of the freedom of the user. I see myself as a user, you see yourself as a proprietary distributor. BSD code helps build DRMed jails for Sony and Apple. I prefer a license that disallows this. I want to share my code with anyone in the world who does not to build proprietary stuff (for building new software). And share my software with anyone for usage, regardless of the purpose.

Re:can it build the linux kernel?

[NatasRevol]

So staying open source is being a bully now?
Vastly improving a product is just 'adding ketchup' now?

Moron.

Re:can it build the linux kernel?

[marcello_dl]

And that is one of the worst justifications ever. In fact, it's so bad, it lends credence to the "GPL is viral" belief - because the GPL can take and take and take, and not give back

it takes what IT IS ALLOWED to take, it gives back everything as long as you redistribute it keeping it free.

There are two civilizations, the Babbling Software Developers (BSD) and the Grooming Prevented by Law (GPL). The first has a law: "do what you want". The other has a law "Do what you want as long as it doesn't prevent others to do the same". Simple, no?

I don't see why a dev that allows everybody, commercial entities wielding software patents included, the freedom to use his code would be bothered by the GPL, but then, I am not one of them.

Re:can it build the linux kernel?

[Anonymous Coward]

Removing GPL licensed code from FreeBSD is a truly good thing.

Re:can it build the linux kernel?

[kthreadd]

There's no spirit of FSF in Linux. Linux has never been associated with FSF in any way. The reason why the FSF is so keen on calling it GNU/Linux is because they failed with their own kernel project.

Re:can it build the linux kernel?

[Anonymous Coward]

Feel free to shave your neckbeard anytime.

Re: can it build the linux kernel?

[Anonymous Coward]

Tomato tomato, the engine is open source and that was what Apple forked anyway.

Re:can it build the linux kernel?

[Anonymous Coward]

And that is one of the worst justifications ever. In fact, it's so bad, it lends credence to the "GPL is viral" belief - because the GPL can take and take and take, and not give back

it takes what IT IS ALLOWED to take, it gives back everything as long as you redistribute it keeping it free.

No. If I add 1 file of GPL code to my BSD project, then redistribute the combination under the BSD (which is free, even according to the FSF), instead of changing the licensing of this 90% BSD project to GPL, I'm satisfying what you just said, but I'm violating the GPL. Yes, I can get around it by licensing the remaining files as BSD, but the whole project with GPV, but the fact remains that the requirements are rather more strict than you're portraying. Yes, they need to be more strict to achieve the intended goal (preventing anyone, anywhere, ever from daring to distribute a binary built from the GPL'ed code without distributing or offering to distribute that GPL'ed code as well), but that doesn't excuse misrepresenting the GPL's conditions as less than they are.

Re:can it build the linux kernel?

[Anonymous Coward]

They are getting close to building the Linux kernel according to these slides.

These slides? Which slides? Your link cannot be understood without hovering over it with mouse.

Re:can it build the linux kernel?

[Anonymous Coward]

Can LLVM's Clang build the Linux kernel and QT/KDE?

I would love to have a Linux system,(not android), completely free of GNU.

http://llvm.linuxfoundation.org/index.php/Main_Page

Re:can it build the linux kernel?

[Anonymous Coward]

You are certainly free to share your code, or not share it, as you please. And everyone else should have that same freedom -- to share it or not share it as they like. Where's the problem?

BSD think of the freedom of the distributor, the GPL thinks of the freedom of the user. I see myself as a user, you see yourself as a proprietary distributor. BSD code helps build DRMed jails for Sony and Apple. I prefer a license that disallows this. I want to share my code with anyone in the world who does not to build proprietary stuff (for building new software). And share my software with anyone for usage, regardless of the purpose.

Re:can it build the linux kernel?

[jeremyp]

You can't change a source code licence simply by compiling the code it applies to with a different compiler.

This, incidentally, is how we know that the fact that FreeBSD is stuck on gcc 4.2.1 has nothing to do with licensing. They could compile their entire code base with gcc 4.million and not infect it with GPL v3 (if they are using a modern version of glibc, there might be an issue, however). The people at FreeBSD are simply objecting to the anti-Tivoisation clause in GPL v3 as is their right. In fact, they object to the whole "viral" nature of GPL version x and presumably would have replaced the compiler a long time ago if there had been a credible alternative.

Re:can it build the linux kernel?

[jeremyp]

You can get the open source WebKit browser which is what Safari and Chrome (until they forked it recently) was based on. What's the problem?

Re:can it build the linux kernel?

[Anonymous Coward]

No, i havent since Nixon was President, and i'll be damned if i do before he is President again...

  -- RMS

Re:can it build the linux kernel?

[Anonymous Coward]

Are you not familiar with links? Welcome to the web, friend. It's UHHHMAZING. I think you'll have fun.

Re:can it build the linux kernel?

[Anonymous Coward]

Feel free to grow one. In a few years. If you ever grow up, kid.

Re:can it build the linux kernel?

[Anonymous Coward]

Wearing closed-toe shoes makes the world a better place.

Re:can it build the linux kernel?

[Anonymous Coward]

I don't see why a dev that allows everybody, commercial entities wielding software patents included, the freedom to use his code would be bothered by the GPL

I think, mostly, that BSD guys don't care if other guys want to be GPL guys. It would be more convenient for BSD guys if all software was BSD-licensed so everyone could grab source code they needed, but nobody expects that.

I did read of a case though where some GPL zealots were being jerks about it. There was a big project under BSD, and the GPL zealots set up a GPL-licensed project explicitly to compete with the BSD project. Every time a patch went in to the BSD code, they grabbed it and stuck it in their GPL code base; but they also made a point of sticking in new features in the GPL code base and said "neener neener neener we have stuff you can't have and all your code are belong to us!" (Not exact words, but there were some really obnoxious emails gloating with this idea, sent to a mailing list.)

The BSD guys were kind of bothered by this, and can you really blame them?

Re:can it build the linux kernel?

[unixisc]

Precisely. Even Linus is on record as saying that he avoids associating w/ 'Free Software' these days due to the divisiveness of that project

Too bad the FSF guys lack the competence to complete their own kernel, while even a university professor in Holland managed to create a complete microkernel based OS that's completely functional

Re:can it build the linux kernel?

[Anonymous Coward]

He could combine a linux kernel with a bsd userland. But he would have to do it himself...

Re:can it build the linux kernel?

[Guy Harris]

Apple is the bully who took your 3 lunch sandwiches, added ketchup to one of them and gave it back to you and made you say "thank you sir for the ketchup".

If you're talking about KHTML, and the original complaints about how Apple treated it, I think it's more like "made copies of the recipes for your 3 lunch sandwiches, changed one of the recipes to include ketchup, and handed you back the raw new recipes without change bars". It's not as if they "took" KHTML in such a way that it immediately ceased to work in KDE.

Re:can it build the linux kernel?

[imnotanumber]

GNU the operating system, not GNU the public license...

Just an extra detail: the G in GPL means "GENERAL" as in GENERAL PUBLIC LICENSE.

Re:can it build the linux kernel?

[tlhIngan]

They broke it of from the community. They made changes but they did not commit the diffs. This after huge pressure from KDE devs. I prefere to have my code eaten up by other GPL projects and remain free forever than get assimilated in Apple's pretty iShit.

They couldn't do that, because KHTML is LGPL.

In fact, Apple released their changes as required by the LGPL.

What KDE devs wanted was better documentation on the changes because Apple's changes were being sent to the KHTML guys as large diffs without much commentary. As WebKit and KHTML diverged, the pages became more and more difficult to integrate, and there's little information as to what changes were to improve things and others were bug fixes.

That was the main beef with the KDE guys - Apple was releasing source code as required, but they couldn't be integrated.

There was never any "code lockup" and in fact, Apple was providing the KHTML guys with patches out of kindness (remember, the (L)GPL never states you need to give the changes back upstream - you're free to fork it as you wish).

The code was always free. It's the upstream guys were getting patches that they couldn't use was the problem, and Apple has every right to ignore them if they wanted.

What Apple did in the end was put up a public CVS server with all the changes, so the KDE folks could see the diffs as smaller patches as well as individual change logs. Neither of which is a requirement of the GPL.

Otherwise RedHat would be in trouble because they're doing the same thing with Oracle - they're combining their patches into one huge patch so guys like Oracle can't pick and choose easily.

WebKit is still LGPL, by the way. It's never been anything other.

Re:can it build the linux kernel?

[Bengie]

Ohh, thanks, so I can put my code under any license I want, but my code is worthless with out the other code.

I'm not trying to make an argument about entitlement, just that it's a moot argument that reads like "If you don't like your food, you're free to not eat it and die".

Most people will keep eating the food because it's better than not eating, but at some point, they might get... fed up... lawlawlawlawl. Maybe they like the food, no problems. Claiming GPL is free is like claiming Verizon actually sells unlimited FIOS. For all intents and purposes, it's free, but there are cases where it blows up in your face.

Re:can it build the linux kernel?

[Bengie]

"Do what you want as long as it doesn't prevent others to do the same"

Not quite. BSD also doesn't allow other to prevent you from doing the same. GPL prevents you from, in their words, "stealing" code, and making your own changes and not sharing YOUR changes. They force you to share your changes, for the good of the user of course.

Most people love sharing, but being forced to share is a whole other issue. I don't mind giving money to charity, but if those charities walked into my house and took a dollar from my pocket, that's a bit different. Not a great analogy, but it conveys my "feelings".

If you don't like it, you don't have to use it, right? Well, the vocal group in the GPL community are like zealots and constantly spread FUD about BSD and how people can steal your code! omg! my code is no longer in version control! Someone must have stole it! The only thing wrong with GPL is some of their evangelists and the fanaticism of the ideology.

I get this whole, holier than thou, kind of vibe.

Re:can it build the linux kernel?

[Anonymous Coward]

The kernel is not GNU, because GNU is not a license. It is definitely GPL.

https://www.kernel.org/pub/linux/kernel/COPYING

Re:can it build the linux kernel?

[marcello_dl]

> They force you to share your changes

Citation needed. IIRC You can take gpl and do what you want with it in your organization, IF you don't redistribute the code. Hosting code and let users be clients is the same as distributing the code in GPLv3, and it makes a lot of sense, because the net result is the same of a distribution.

Re:can it build the linux kernel?

[Bengie]

Yes, GPL works if you don't want to share anything, but if you want to share something but not everything, it won't work. GPL is an all-or-nothing license.

"Perfect is the enemy of good"

"Of all tyrannies a tyranny sincerely exercised for the good of its victim may be the most oppressive. ..."

"The road to hell is paved with good intentions" - Not to mean that GPL is bad, just that it has many unforeseen consequences to it's practicality.

Re:can it build the linux kernel?

[marcello_dl]

You are not simply sharing something, you are building upon other people's code. If you bear this in mind, the tyranny begins to belong more to the guy closing down stuff, IMHO.

If you want to keep parts proprietary, build upon different people's code instead. Seems not tyrannical at all.

Re:can it build the linux kernel?

[TheRaven64]

Why would you point to the tarball location, when the revision control system is online?

More interesting page for FreeBSD 10...

[Noryungi]

Try this one: https://wiki.freebsd.org/WhatsNew/FreeBSD10

I have to say, this is shaping up to be a very interesting release. Bhyve, in particular, sounds very promising...

Re:More interesting page for FreeBSD 10...

[Anonymous Coward]

Bhyve, in particular, sounds very promising...

And it has quite a catchy name!

Re:More interesting page for FreeBSD 10...

[MatthiasF]

I think the updates to ZFS are more impressive. The addition of LZ4 compression (50-80% faster), SSD TRIM support and the new NOP-write that reduces the need to write to compressed disks will all mean commodity hardware SANs running FreeBSD will be able to give the big boys a run for their money.

So, better compression, SSD speed-up and less need to write to changed files, means there will be a huge performance increase.

Re:More interesting page for FreeBSD 10...

[MtHuurne]

Looking at the logo on their home page, I think it's pronounced "bee hive".

Re:More interesting page for FreeBSD 10...

[SuricouRaven]

Can we take drives out yet? ZFS makes it nice and easy to just add a new drive to a storage pool, but taking one out (ie, freeing up port for a larger drive, or turning your 16x1TB drives into 8x2TB drives to reduce the frequency of failures) is a different matter.

Re:More interesting page for FreeBSD 10...

[Bengie]

Not yet, that requires pointer re-writes, which has been talked about for a long time by the ZFS devs.

Essentially, ZFS does not currently allow data to be moved once written. Supports Delete and Insert, but no Update. A big reason for this is if you move the data, you need to update all points that address that data, and to do so in an atomic way that doesn't leave the system in an unstable state.

Re:More interesting page for FreeBSD 10...

[MatthiasF]

I do not think there is a safe way to do that in regards to any storage array system, besides setting up the two arrays in parallel and copying the data.

ZFS will allow you to migrate up to larger drives, but I believe it has to be the same number of drives and it is still fairly risky (requiring you to degrade the array for a rather lengthy period).

Here's a blog explaining some.

https://www.dan.me.uk/blog/2012/11/14/increase-capacity-of-freebsd-zfs-array-by-replacing-disks/

I think you can also use the mirror capabilities to migrate to a newer set of drives, but I have not seen anyone explain the process well.

Re:More interesting page for FreeBSD 10...

[Anonymous Coward]

Pretty sure you can make a Dell PERC Raid 5 volume larger by replacing the drives one at a time, expanding the virtual drive and then the volume (if NTFS, can do on the fly.)

Equallogic lets you replace drives my using a dedicated hot spare as a mirror drive and you could foreseeable replace all drives in this manner. You can do things like convert from RAID 50 to RAID 6 on the fly. I'm sure other vendors have similar capabilities.

Re:More interesting page for FreeBSD 10...

[daboochmeister]

With all due respect to your low /. id#, why do you say that? A type 2 hypervisor - are there no good type 2 hv offerings on BSD today? Are there specific production uses for FreeBSD that you think have been forestalled to date, that this opens up? Or is there something about bhyve that sets it apart from the existing options?

Re:More interesting page for FreeBSD 10...

[SIGBUS]

The best way to rearrange a pool is to create the new pool with the desired disks, and then use zfs send to send the data from the old pool to the new one. Even if block pointer rewrites were implemented, I'd still be more trusting of a send (besides, that means you have a backup if everything goes pear-shaped).

Re:More interesting page for FreeBSD 10...

[Anonymous Coward]

Or is there something about bhyve that sets it apart from the existing options?

bhyve is about 1/10th the code and resulting size of almost all other VMs, while still maintaining most of the features.

bhyve is also integrated into the Kernel and UserLand, which gives some interesting abilities that complement ZFS. It is only a few simple commands to snapshot and boot a bhyve VM. You can effectively treat your OS like a database; begin transactions and easily roll-back.

Can you imagine

[Anonymous Coward]

A world where we can call it Linux and the nerd who immediately corrects you with GNU/Linux will have to shut up for once?

Re:Can you imagine

[wagnerrp]

Well there is Android...

Re: Can you imagine

[Anonymous Coward]

GNU/Android if I may

Re:Can you imagine

[Anonymous Coward]

And various embedded systems that run a non-GNU libc and userland on top of a linux kernel.

Re:Can you imagine

[Anonymous Coward]

Technically it's called FreeBSD. ;)

Re: Can you imagine

[Anonymous Coward]

There isn't a single line of GNU code in Android.

Re:Can you imagine

[Hypotensive]

I can imagine a world where an AC posts shit flamebait in the same thread multiple times trying to pretend that he's different people.

Re: Can you imagine

[wagnerrp]

If you want to be that way, it would technically be Android/Linux.

Re: Can you imagine

[Anonymous Coward]

Seek and ye shall find

https://f-droid.org/repository/browse/?fdid=info.guardianproject.lildebi

Re: Can you imagine

[marcello_dl]

In fact, given all those android installations, those who call GNU/Linux "Linux" are now less correct than those calling it "GNU".

Re: Can you imagine

[Darinbob]

Or is that Android/GNU/Linux?

Re: Can you imagine

[wagnerrp]

Huh? It's either the Android OS or the GNU OS. Not both. Are you suggesting some abominable combination of the two?

Netcraft confirms it

[Megane]

It is now official. Netcraft has confirmed: GCC is dying One more crippling bombshell hit the already beleaguered GCC community when IDC confirmed that GCC market share has dropped yet again, now down to less than a fraction of 1 percent of all servers. Coming on the heels of a recent Netcraft survey which plainly states that GCC has lost more market share, this news serves to reinforce what we've known all along. GCC is collapsing in complete disarray, as fittingly exemplified by failing dead last in the recent Sys Admin comprehensive networking test.

You don't need to be the Amazing Kreskin to predict GCC's future. The hand writing is on the wall: GCC faces a bleak future. In fact there won't be any future at all for GCC because GCC is dying. Things are looking very bad for GCC. As many of us are already aware, GCC continues to lose market share. Red ink flows like a river of blood.

GCC is most endangered of all in FreeBSD, having lost 93% of its core developers. The sudden and unpleasant departures of long time FreeBSD developers Jordan Hubbard and Mike Smith only serve to underscore the point more clearly. There can no longer be any doubt: GCC is dying.

Let's keep to the facts and look at the numbers.

OpenBSD leader Theo states that there are 7000 users of GCC. How many users of NetBSD are there? Let's see. The number of OpenBSD versus NetBSD posts on Usenet is roughly in ratio of 5 to 1. Therefore there are about 7000/5 = 1400 NetBSD users. BSD/OS posts on Usenet are about half of the volume of NetBSD posts. Therefore there are about 700 users of BSD/OS. A recent article put FreeBSD at about 80 percent of the GCC market. Therefore there are (7000+1400+700)*4 = 36400 FreeBSD users. This is consistent with the number of FreeBSD Usenet posts.

Due to the troubles of Walnut Creek, abysmal sales and so on, FreeBSD went out of business and was taken over by BSDI who sell another troubled OS. Now BSDI is also dead, its corpse turned over to yet another charnel house.

All major surveys show that GCC has steadily declined in market share. GCC is very sick and its long term survival prospects are very dim. If GCC is to survive at all it will be among OS dilettante dabblers. GCC continues to decay. Nothing short of a miracle could save it at this point in time. For all practical purposes, GCC is dead.

Re:Netcraft confirms it

[Anonymous Coward]

FFS if you're going to use that old copypasta, at least make an effort to at least superficially seem coherent and relevant.

I mean stuff like

Due to the troubles of Walnut Creek, abysmal sales and so on, FreeBSD went out of business and was taken over by BSDI who sell another troubled OS. Now BSDI is also dead, its corpse turned over to yet another charnel house."

BSDI? Walnut Creek? OS? You're not even trying.
Score: "-1, Me_Too"

In related news in the real world nobody gave a shit what compiler FreeBSD uses, since it's just used by autistic die-hards who refuses to get on with the program these days.

Re:Netcraft confirms it

[Anonymous Coward]

Jim Clapper, is that you?

Re:Netcraft confirms it

[interval1066]

OpenBSD leader Theo states that there are 7000 users of GCC.

I'd like to see how de Raadt came up with THAT number. Seems very low to me. Accross the world? Or is that a blanket, unqualified statement? 7K users of GCC? Come on...

Re:Netcraft confirms it

[Anonymous Coward]

WOOSH!

Re:Netcraft confirms it

[Anonymous Coward]

Are you fucking shitting me?

Re:Netcraft confirms it

[Anonymous Coward]

OpenBSD leader Theo states that there are 7000 users of GCC.

I'd like to see how de Raadt came up with THAT number. Seems very low to me. Accross the world? Or is that a blanket, unqualified statement? 7K users of GCC? Come on...

Seriously?

YHBT. YHL. HAND.

Re:Netcraft confirms it

[gagol]

In that case, I think you CAN shoot the messenger...

Re:I'd be suspicious

[kthreadd]

By inspecting the source code.

Clang is Slower

[tuppe666]

For many reasons.

Clang is faster...

No its slower. Phoronix benchmark GCC vs Clang all the time. http://www.phoronix.com/scan.php?page=article&item=llvm_clang33_3way&num=1

Re:Clang is Slower

[Anonymous Coward]

No its slower. Phoronix benchmark GCC vs Clang all the time. http://www.phoronix.com/scan.php?page=article&item=llvm_clang33_3way&num=1 [phoronix.com]

Interesting that they use -O3, which is not clang's fastest optimisation setting (which is -Ofast). Also, they are comparing the speed of the compiled code (which was my second point, that it produces faster code), not the speed of the compiler. Clang is currently (in my brief tests) roughly 4 times faster (as in, the compiler's exception time is 4 times faster) than gcc at compiling C, and 2 times faster than gcc at compiling C++ when set to -O0.

Re:Clang is Slower

[gnasher719]

No its slower. Phoronix benchmark GCC vs Clang all the time.

The question is: Do GCC and Clang benchmark Phoronix?

As a long time Clang user, I never heard of Phoronix. Maybe it's different with gcc users. Maybe gcc develops use this Phoronix benchmark to drive their software and optimisation development.

Re:Clang is Slower

[Anonymous Coward]

Phoronix are also stunningly incompetent and it's a mystery why anybody would listen to a damn thing they say.

Re:Clang is Slower

[gcore]

Considering how bad the articles and journalism is, perhaps you shouldn't put too much faith in the benchmarks of Phoronix.

Re:Clang is Slower

[UnknowingFool]

Define "all the time". In the article you linked, GCC was slightly faster in most tests than Clang especially when it came to OpenMP however it didn't beat it in all tests. The performance gaps in most tests were small except when it came to OpenMP which Clang does not support.

Re:Clang is Slower

[Anonymous Coward]

Actually, my bad, they do benchmark the speed of the compiler there as well as the speed of the code it produces. Their results are that clang runs significantly faster. You're wrong, and your own benchmark even says you are.

Re:Clang is Slower

[Anonymous Coward]

The performance gaps in most tests were small except when it came to OpenMP which Clang does not support.

Yet

Re:Clang is Slower

[tlhIngan]

No its slower. Phoronix benchmark GCC vs Clang all the time. http://www.phoronix.com/scan.php?page=article&item=llvm_clang33_3way&num=1

Except FreeBSD is not using GCC 4.8. They're using GCC 4.2.1, the last version that was GPLv2.

While the current version of GCC may be faster than Clang/LLVM, that doesn't mean Clang/LLVM isn't faster than what is in use now, so the switch may even boost performance compared to the ancient version of GCC in use.

Re:Clang is Slower

[Anonymous Coward]

Point out the part where clang is faster. I've just looked at the linked benchmarks and clang doesn't appear to be "significantly" faster to me.

Re:Clang is Slower

[Anonymous Coward]

No, it's so much better to just trust some user sprouting off on slashdot. Much more credible.

Re:Clang is Slower

[hedwards]

It's not just about faster, compilers sometimes have stability and reliability problems when using higher levels of optimization. GCC wasn't able to handle much more than -O2, IIRC, and even at -O2, there were warnings about not bothering to report bugs without first trying it with lower levels of optimization.

If clang is able to handle -O3 in a production environment without the warnings, that's perfectly legitimate to compare with gcc with current default optimizations.

Re:Clang is Slower

[hedwards]

Right and it's a moot point if Compiler X is faster than both of them if the license requires that you be riding a unicycle while compiling.

I know a lot of the Linux zealots don't care to acknowledge it, but the reality is that the viral clauses are just a matter of cutting off your own nose to spite your face. There's a real incentive to contributing back patches in that it's expensive to maintain your own fork, just for a few patches.

Re:Clang is Slower

[adam.vany]

The GPL isn't about providing incentive to contribute back patches (or forcing it as some characterize it). It's about ensuring software freedom for software *users*. That any software you use, you should be allowed to study it, modify it, and re-distribute it without restriction. Economic benefits from open source are not what the GPL is for, so don't claim that the GPL is unnecessary because companies will still contribute back. That's a mis-representation of what Free Software is all about. If you want to complain about "zealots" at least understand their position.

Re:Clang is Slower

[Anonymous Coward]

I understand the position. You apparently don't.

The GPL is about forcing people to opensource the code they write if they wish to use GPL code. You can gussy it up however you like, but at the end of the day, you're telling the users of the code that they can't do somethings that they would otherwise be able to do under a permissive license.

All this software freedom bullshit is just ridiculous. *BSD has proven over the last couple decades that you don't have to infringe upon people's ability to use the code in order to ensure that people give back and don't take away from the community.

With the lastest version of the GPL, they've taken the step of specifically making it challenging to use the code on proprietary projects, even when previous versions of the license would have been just fine.

Seems like zealotry to me.

Re:Clang is Slower

[nullchar]

YOU publish your own reproduce-able (and tweak-able, if you don't like their chosen settings such as -O3) benchmarks, and I'll be happy to look at them instead of Phoronix.

Re:Clang is Slower

[Anonymous Coward]

The system code isn't that likely to be using OpenMP, and for anything else clang and gcc are so close it doesn't really matter. The only case where it's important is for the programs in the port collection that do use openMP to good effect - and those can be compiled with gcc (even a completely new gcc, from its own port) quite easily. It can even be set to happen by default.

Re:Clang is Slower

[Anonymous Coward]

Page 3 –the only page on which they compare the speed of the compiler, not the speed of the produced code. Clang comes out literally twice as fast.

Re:Clang is Slower

[Anonymous Coward]

While I haven't done a full regression test on it, clang with both Os and O3 would miscompile all sorts of shit from 2.9 to 3.2 (the earlier versions were 'when it would compile at all'). As such I'm still a little wary of a full fledged jump, but given both how long in the tooth 4.2 is getting combined with all the bugs across gcc versions, there no reason *NOT* to switch to clang, and at least on the bright side they might get a more bugs fixed in it without introducing new ones (unlike gcc).

The other perk, although I might be wrong: clang has far fewer library dependancies than gcc ATM. llvm+clang+libcxx+libunwind+glibc, plus one of the cxx abis, and afaik that's it. gcc at the moment requires gmp, mpfr mpc, and a whole bunch of other shit if you want the graphite extensions. Combined with it no longer being c-compilable (not that it ever seemed to compile right on a non-gnu bootstrap to begin with...), there seem fewer and fewer reasons to use gcc over the alternatives.

Re:Clang is Slower

[smash]

Phoronix - adview trolling website for Linux/GNU fanboys to set up rigged benchmarks of Linux vs. X, to get users/developers/fans of X enraged and visiting the site.

The number of massively flawed benchmarks done by phoronix lead me to believe this is the only logical purpose of the site. One or two benchmark screw ups is one thing, but to consistently screw up benchmarks to make everybody else's software look bad via subtle omission ro selection of bad configuration is just trolling.

Re:Clang is Slower

[smash]

I know a lot of the Linux zealots don't care to acknowledge it, but the reality is that the viral clauses are just a matter of cutting off your own nose to spite your face. There's a real incentive to contributing back patches in that it's expensive to maintain your own fork, just for a few patches.

Let alone going to the trouble of writing your own entire compiler framework...

Re:Clang is Slower

[Anonymous Coward]

No. The GPL is about preventing others' profiting from your work and providing you with zero compensation. If you want to use my code in your closed-source product that you're selling for $500-per-seat, you can damn well pay me to grant you usage under a different license. If you don't want to buy a license from me, then you can choose to release your own product also under GPL.

Re:Clang is Slower

[Bengie]

Users never study/modify code, only developers do and GPL claims to not care about them, only users. I stand behind GPL's ideology, but it is not as practical as BSD in the real world. I do hope for the day when we won't need BSD anymore and GPL will work for all cases, but I don't see that day happening until Software Patents and DRM is gone.

Good move

[Anonymous Coward]

I'm glad for this move for technical reasons rather than philosophical. Clang produces better warnings and error messages. It is also more fault tolerant and faster. I have almost completely moved to using Clang instead of GCC for my open source projects. It's just a nicer compiler to work with.

Not sure how much of the problem is code and how much is the compiler, but I've found at least one application that will crash when compiled with GCC, but will run smoothly when compiled with Clang. My guess would be there is a code issue involved, or an optimization issue, but I still find it interesting that two compilers passed the same flags produce such different code that one binary works flawlessly and the other regularly crashes.

Re:Good move

[bluefoxlucid]

It's faster because it doesn't support as much optimization. Some analysis simply isn't implemented. That isn't to say it won't be faster when it's on even footing; just that right now huge swaths of algorithms are missing, so the output code doesn't perform as well but it does get produced in much less time.

Re:Good move

[Anonymous Coward]

It actually is somewhat "philosophical."

Due to disagreements with the GPL, especially with revision 3.0 and onwards, they've been shipping GCC 4.2.1 to this day.
It's no surprise that clang would compare far more favourably in terms of bugs, features, and performance, especially when you consider that the former is a relic from at least 6 years ago.

LLVM and GCC have both made enormous strides in that time.

Re:Good move

[Anonymous Coward]

Actually GCC sucks *because* of its license. All the *best* developers DESPISE the GPL and see right through its statist objectives, so they have largely switched to LLVM. Very few top-quality apps these days are compiled with GCC because no one wants to deal with the socialist/statist idiots who push that garbage.

Re:Good move

[gagol]

I will take a compiler that optimize less and make sure code is correct against one that try roo hard and break the software.

Re:Good move

[bluefoxlucid]

Not mutually exclusive. The algorithms can be implemented correctly, but will still take more time. The output will be of better quality, of course.

GCC suffers a lot from trying to change the underlying behavior into a static single assignment model when it was already mature. This is a good model, easy to analyze and ensure correctness; but rewriting a hell of a lot of complex code that does complex analysis to instead do the same analysis on a different type of data can result in errors. Initial implementation means understanding the internal representation, understanding the optimization, and understanding manipulation; re-implementation means understanding the old and new internal representations, understanding the optimization, understanding how the optimization is implemented, and understanding the old and new manipulation, and then figure out how to rewrite one piece of code to bring in different data and operate the same logic on it and then perform the same manipulation via a different mechanism.

When you think about it, GCC is a really hard core piece of software and showcases some excellent work. Clang will probably be better quality because they're doing it the best way to start with, and so they can focus on getting it correct the first time instead of trying to rewrite existing correct code into new code that uses a different mechanism without breaking it.

Re:Good move

[gnasher719]

It's faster because it doesn't support as much optimization. Some analysis simply isn't implemented. That isn't to say it won't be faster when it's on even footing; just that right now huge swaths of algorithms are missing, so the output code doesn't perform as well but it does get produced in much less time.

On the other hand, the Clang static analyzer alone is worth burying gcc in a big, big hole and leaving it there forever.

Re:Good move

[david_thornley]

Optimizers try to keep correct code correct, and in my experience do a good job of it. If a program works with Clang, and doesn't on g++ with optimization turned up, I'd suspect there was undefined behavior in the program.

There's tradeoffs here. An optimizer that assumes it's working on correct code can do better, but C and C++ have a lot of undefined behavior, and in some circumstances a compiler that doesn't optimize as aggressively and accepts some undefined behavior may be more useful. On the third hand, you don't know if the less aggressive optimizer is interpreting the undefined code the way you'd expect, so there's advantages to simply and obviously failing.

Re:Good move

[TheRaven64]

With my LLVM dev hat on, please can you point to some analyses that are implemented in GCC and not LLVM? I have some masters' students in need of interesting projects in a few months...

Re:Good move

[m.dillon]

GCC tends to overoptimize things to the point where the optimizations become self-defeating.

-Matt

on Single Compilers

[bill_mcgonigle]

I don't think I've seen Ken Thompson's Reflections on Trusting Trust come up more in the past decade than it has in the past week. Right now seems like a particularly inauspicious time to switch to a one-compiler-to-rule-them-all strategy.

I'm particularly interested in trying to build the gcc phase-1 bootstrap compiler with llvm to see how that works out (TODO list...).

Re:on Single Compilers

[phoenix_rizzen]

FreeBSD (and Linux) were already in a "one compiler to rule them all" situation, aka GCC. At least with FreeBSD 9 there were two compilers in the base install (GCC, LLVM), and you have the option of keeping GCC in 10 if you really want.

IOW, you're complaint is baseless and backwards.

Re:on Single Compilers

[Anonymous Coward]

Trust. Motherfucker.
http://www.dwheeler.com/trusting-trust/

Trust that.

Re:on Single Compilers

[steelfood]

If the theoretical exploit isn't at the hardware level, one way to discover a compromised compiler would be to compile the same code with multiple compilers, and then compare their outputs. Having multiple compilers for one platform is a good thing.

Re:on Single Compilers

[Eravnrekaree]

You will have the option of using GCC, its in the packages system and you can easily install it. Clang is now the default, but of course GCC can still be used if you wish.

Re:on Single Compilers

[bill_mcgonigle]

At least with FreeBSD 9 there were two compilers in the base install (GCC, LLVM), and you have the option of keeping GCC in 10 if you really want.

Right, FreeBSD 9 was doing it the best way, and now FreeBSD 10 is doing it a worse way.

IOW, you're complaint is baseless and backwards.

If my complaint is "removing additional compilers from the base install will discourage their use", please describe how the complaint is baseless. Do you feel that having multiple compilers per platform is not an advantage for the platform's security posture? Please explain how that assertion is backwards, specifically in regards to the issue of trusting compilers.

Re:on Single Compilers

[Anonymous Coward]

And they will both differ, with different trojan horses in both of them.
So then you need a third compiler, and then a fourth one, just in case (tm)...

Re:on Single Compilers

[Anonymous Coward]

FreeBSD's GCC is version 4.2 and completely unsupported. Why would you want FreeBSD to carry forward a compiler nobody uses or wants?

Re:on Single Compilers

[gmueckl]

You should read Ken Thompson's "Reflections on Trusting Trust" and the thesis that shows how to detect that attack (forgot the name of the author - Bruce Schneier has written a good summary of it, though). This will tell you why you want multiple compilers for some things.

Very short summary (probably wrong - it's confusing): You want to compile your production compiler with two different compilers to check if the binaries these resulting production compiler binaries produces are the same (e.g. you let the production compiler compile itself again using the two different binaries). Your alternate compiler does not strictly have to be up to date or fully featured for that. It must only be able to compile the compiler that was used for creating the actual product under scrutiny.

Re:on Single Compilers

[Anonymous Coward]

10 as of today defaults to gcc on some platforms, and clang on some. It does not build the non-default compiler by default, but that can be toggled easily. It will compile and run with gcc on all supported platforms, and it will compile and run with clang on some (there are a few platforms where clang kind of works but isn't default).

I don't really see the problem with this?

Re:on Single Compilers

[evilviper]

If my complaint is "removing additional compilers from the base install will discourage their use", please describe how the complaint is baseless.

Encouraging the use of GCC won't solve the "trusting trust" problem on its own... Installing from ports/pkgs is trivial... There is still a huge installed base of GCC compilers out there. It's not as if FreeBSD just assassinated the GCC upstream. GCC has a HUGE base, and encouraging use and development of LLVM is the move that will get healthier competition in the open source compiler space.

Re:on Single Compilers

[serviscope_minor]

Massive meh.

GCC still has a better optimizer than LLVM, sometimes by quite a wide margin. So unless you recompile everything, kernel included with GCC, you lose out performance wise.

Re:on Single Compilers

[Anonymous Coward]

Ssa (cis) er'uoy etib ti fi neve tenmmoc nettirw sdrawkcab a ezingocer ton dluow uoy.

Re:on Single Compilers

[david_thornley]

Compiler output will differ between compilers, but the compiled code should do the same thing. So you compile a compiler with multiple compilers, and use your newly compiled compilers to compile the same compiler, and the second-stage compilers should be pretty well identical.

Re:on Single Compilers

[MechaStreisand]

Not even close to true in FreeBSD's case: the version of GCC they use is 4.2, and LLVM is much better than that.

Re:I'd be suspicious

[philipmather]

Redundant, keep reading the whole "What's New" page and the conspiracy klaxon attached to your tin-foil hat will go into hyper-overdrive...

"
Support for the RDRAND random number generator

Status:
Committed to -CURRENT, MFC-ed to 9-stable
Author:
Konstantin Belousov
Web:
http://svn.freebsd.org/changeset/base/240135
RDRAND is the new Intel's CPU instruction for accessing its hardware random number generator, also known as the code-name Bull Mountain. It is present in Ivy Bridge and newer CPUs.
" ...although it would seem more likely that Konstantin Belousov would be working for the FSB rather than the NSA really. ;^)

Re:Goodbye statist GPL don't let the door hit ya!

[Anonymous Coward]

FreeBSD is looking set to kick some serious Linux ass!

Yeah, let me know how that goes!

Re:I'd be suspicious

[dkleinsc]

That is insufficient, as Ken Thompson demonstrated decades ago.

Re:I'd be suspicious

[Anonymous Coward]

How do we know that LLVM/Clang hasn't already been backdoored by the NSA?

How do we know your mom wasn't backdoored? There could be NSA DNA in you right now!

two dictionaries say "no such word"

[raymorris]

The two dictionaries I checked, including Merriam-Webster, don't list "performant" as a word.

Trying so hard to use "big words" that you resort to using non-words that sound big = douche. Of course "douche" is actually a word.

Re:I'd be suspicious

[interval1066]

Support for the RDRAND random number generator

Now you can go "WHY????"

Re:Goodbye statist GPL don't let the door hit ya!

[Anonymous Coward]

FreeBSD: Core Routers, Netflix, Verisign, ZFS
Linux: Steam, WordPress, I think my data is correct


/Troll

Re:two dictionaries say "no such word"

[rubycodez]

it's origin is Microsoft wank speak.

It is a French word, maybe that Microsoft weenie also said things were tres this or that. eurotrash wannabe.

Why?

[Anonymous Coward]

"Especially GPLv3 which they don't want to include in their base OS."

Why?

"GPLv2 was never a problem."

Why is the GPL3 a problem, then?

Re:Why?

[jonwil]

People are using BSD (and BSD code) in devices and products (consumer products and otherwise) where there is a requirement (might be a requirement specified by some law or regulation, might be a requirement specified by a content provider or other entity, might be a business requirement, might be something else) to ensure the software on the device can't be changed by the end user. And using a gplv3 version of GCC (and libstdc++/libgcc/etc) creates a conflict between the anti-tivoization clauses in gplv3 and the requirements of the software developers to prevent tampering with the software.

Re:Why?

[Anonymous Coward]

I'm seeing a lot of "there might be" but none of that is actually possible.

The output of GCC is not GPL'd any more than the output of Microsoft Visual Studio is licensed under the Microsoft Closed Source license.

The output of the GCC cannot cause the code produced to be under the GPL restrictions.

CAN.
NOT.

"And using a gplv3 version of GCC (and libstdc++/libgcc/etc)"

There is not GPL3 libstc/libgcc/etc.

And again, that doesn't any of those "might" to "ensure the software on the device can't be changed by the end user." cannot be applied.

Moreover, those requirements that "might" exist are also illegal restraints on first sale/ownership doctrine/etc of many countries, so by insisting so, the BSD clause is enabling criminal acts.

" the anti-tivoization clauses in gplv3 " DO NOT APPLY to the output of the GCC.

They DO NOT APPLY to the product that merely aggregates the GCC in the product.

Indeed, if the GCC OR ANY COMPILER is included in the product, then how the HELL are you supposed to expect you are going to have restrictions to "ensure the software on the device can't be changed by the end user." WHEN THE ONLY USE FOR INCLUDING A COMPILER IS TO DO THIS???

Again, I ask "Why?"

Your response is nonsensical paranoic nutbar wordsalad.

Appropriate captcha "Cobblers". Precisely what I think.

Re:Why?

[David Gerard]

tl;dr Apple is allergic to the GPL v2, and really allergic to the GPL v3. They were never happy about gcc being their base compiler, so they've been funding the hell out of LLVM/Clang so they can use that as their base compiler instead. Apple is also the main backer of FreeBSD.

Re:Why?

[Cramer]

This would only be an issue if the "consumer device" were shipped with GCC on it. The complied result ("binary") from GCC is not bound by GCC's license -- if that were true, the entire world is violating the GPL.

The simple truth is FreeBSD purists have always had their panties in a wad from anything and everything that wasn't "BSD licensed". The changes in GPLv3 have been enough of bad taste to get the rest of the gang to agree enough is enough.

Re:Why?

[gnasher719]

This would only be an issue if the "consumer device" were shipped with GCC on it. The complied result ("binary") from GCC is not bound by GCC's license -- if that were true, the entire world is violating the GPL.

So what happens with OpenGL or OpenCL, where code is compiled at runtime? That's the real reason why Apple found gcc 4.3 and later unacceptable, and I have some suspicion that was actually a reason for the license change. There are plenty of other places where compilation happens at runtime, like JavaScript.

Re:Why?

[Cramer]

We aren't talking about OpenWhatver, or JavaShit. We're talking about GPLv3 versions of GCC. It's all but unheard of for gcc to be included on a shipping consumer product -- DVR, NAS, router, switch, access point, cellphone, voip phone, pocket watch, coffee maker, network card, and so forth.

FreeBSD has a passable compiler toolchain that isn't beholden to GPL, and they're running with it. Go for them. Someday it might produce code as small, fast, and efficient as gcc. (it you've ever worked with gcc source, you know what it's like to claw your eyeballs out. I don't practice that voodoo anymore.)

Re:Why?

[Anonymous Coward]

s/funding the hell out of/writing/

Clang was a project started at Apple, by Apple employees.

Re:Why?

[Anonymous Coward]

Get the spelling correct. It is appl€ now. Turnabout is fair play.

Re:Why?

[Dog-Cow]

The LLVM toolchain, especially clang, is also way better than gcc for Objective-C code. I think the ease of modification is the true reason that Apple prefers it over gcc.

Re:Why?

[smash]

That. It's also way better for plugging into an IDE, generates way better compiler error messages, etc.

Re:Why?

[Anonymous Coward]

Apple is also the main backer of FreeBSD.

No, no they aren't.

http://www.freebsdfoundation.org/donate/sponsors

Apple only appears here as "Apple Matching Gifts", i.e. Apple's program to match employee contributions to charities. I have to say that it's... interesting that FreeBSD qualifies as a charity for the purposes of that program, but Apple itself does not fund FreeBSD, nor do they help maintain it. (OS X is not rebranded FreeBSD, it's more like a major fork with differences significant enough that it doesn't make sense for Apple to maintain the original or vice versa.)

Also, Apple happily used lots of GPL v2 code over the years, and contributed to v2 projects, and in fact continues to do so -- see for example CUPS. It's only v3 that they've avoided like the plague.

Re:Why?

[TheRaven64]

Lots of our downstream consumers (especially embedded systems companies) have a policy of 'no GPLv3 on the premises'. They don't want the potential liability, and they especially don't want to make it easy for them to accidentally distribute GPLv3 code. Even if they're not planning on shipping GCC in their product (and, by the way, a lot of them do ship toolchains - embedded also includes companies like NetApp and Juniper - as it makes servicing easier), there's the potential to do it accidentally if the code is sitting there on developers' machines.

Re:Why?

[TheRaven64]

Apple doesn't directly fund FreeBSD, but they do sometimes contribute code. For example, most of the MAC framework (SEBSD) was funded by Apple and developed for both FreeBSD and OS X (the pluggable policy stuff in that makes the sandboxing for iOS and Mac App Store apps possible). When Apple wants kernel code written, they often contract work to FreeBSD devs, and are sometimes willing to fund the FreeBSD version at the same time because code that's generic enough to work on both kernels.

Re:I'd be suspicious

[kthreadd]

Not if all you want to do is verifying the source code. If your compiler is compromised then that's your problem.

Different OS, different drama

[Anonymous Coward]

It seems from the changelog that there wasn't much drama merging RDRAND random number generator patches, unlike on Linux. But here's the compiler the problem. Oh well

Very much this.

[Anonymous Coward]

Thank you. Note that I'm very firmly on the GPL side... and I couldn't agree more.

There's room for both free licenses. Let's just get along, shall we?

Ah, the BSD "freedom"

[Anonymous Coward]

Freedom to deny others freedom. That's the only reason for avoiding copyleft.

Re:Ah, the BSD "freedom"

[kthreadd]

The freedom to not offer freedom. No one is forced to use your software.

Re:Ah, the BSD "freedom"

[Anonymous Coward]

"Freedom means you cannot dictate to anyone"

Re:Ah, the BSD "freedom"

[Anonymous Coward]

And, surprisingly, A producing X and giving B the freedom to produce Y from X and not give C freedom Y'1 might result in C having freedom Y'2 that C wouldn't have if B was forced to give freedom Y'1 to C, as B will produce Y, without which neither freedom Y'1 or Y'2 will exist.

The common case of this is that code X produced by A is not useful to group C. B can use X to produce something (Y) that *is* useful to group C. However, B cannot afford to do this unless they get paid, and they are not able to get paid for producing Y without blocking certain freedoms from group C. Note that group B may still contribute some of their changes back to the codebase X under the original license; they'll just not (necessarily) contribute all of them.

There are a ton of cases where the above is true. Crowdfunding don't fix it. For a whole lot of cases, crowdfunding *can't* fix it - there are theoretical reasons why crowdfunding end up being sub-optimal in terms of risk distribution and knowledge.

Re:two dictionaries say "no such word"

[Anonymous Coward]

Did you try a computer jargon dictionary? Formal dictionaries do not add jargon until it is no longer jargon.

'Performant' is not jargon because it is not used to communicate a specific technical meaning within a particular field of knowledge. It is slang: an unrecognised synonym for 'better', used by people who want to make a vaguely positive but non-specific claim. In this case, suggesting that GCC 'produces more performant code' could imply that it produces an output that executes faster, or is smaller, or is a more exact interpretation of the source with respect to the official language standards or a host of other possible claims. In other words, it's equivalent to saying 'produces better code'.

Trying so hard to use "big words" that you resort to using non-words that sound big = douche.

So what you mean to say is, "durr, people who are smarter than me are douches."

Using made-up words is not a sign of intelligence. Smart people know how, when and whether to use the more obscure parts of the English language.

Re:two dictionaries say "no such word"

[Anonymous Coward]

Or maybe his mother tongue is french and this was only a bad translation for "plus performant" which is quite common over here.

Re:two dictionaries say "no such word"

[Anonymous Coward]

dictionaries I checked, including Merriam-Webster,

It's funny how the Zionists presume that the Merriam Webster is in some way authoritative, for our language.

As opposed to the pidgin derivative, with which they are familiar.

Re:two dictionaries say "no such word"

[jones_supa]

The two dictionaries I checked, including Merriam-Webster, don't list "performant" as a word.

What was the other dictionary?

Re: Enough with the xkcd, please.

[UnknownSoldier]

Translation: I'm too lazy to skip comments I don't like and would rather whine about it instead.

Re:two dictionaries say "no such word"

[Lendrick]

Note to self: Using words that aren't words is an excellent way to troll.

I (unknowingly) used something that isn't actually an english word. You took the time to look it up in two separate dictionaries basically just so you could be pedantic. If you're that kind of person, I'm glad I could waste your time. :)

Re:two dictionaries say "no such word"

[Anonymous Coward]

Kind of like the way they use the word "innovation" in every other sentence. After a while people (i.e. PHBs) start to believe it.

(snicker)

Re:Enough with the xkcd, please.

[theskipper]

I feel the same way about ACs.

Then why bother with BSD?

[Anonymous Coward]

That "reason" is no better than "FreeBSD is not using Clang because of licensing" because it isn't using an earlier version of the BSD license that asked for included attribution.

They're entitled not to like it, but that doesn't mean they have a REASON not to use it.

BSD are just pissed off that they don't get the adulation they feel they deserve for sucking the cock of private industry.

And neocons applaud them for the service.

Re:Then why bother with BSD?

[Anonymous Coward]

BSD are just pissed off that they don't get the adulation they feel they deserve for sucking the cock of private industry.

At least it's a clean, nice smelling cock with hairs neatly trimmed around. Sucking Stallman's cock, OTOH, is likely not a pleasant experience, assuming you can find it in all the bush.

Re:Enough with the xkcd, please.

[Zeromous]

Or at least, SOMETHING AWFUL

Re:I'd be suspicious

[chaim79]

However, if you compile the LLVM/Clang in another compiler (say GCC) then you should be good, injecting "compile into" backdoors into all compilers, to target all compilers, would definitely be noticeable.

Re: Enough with the xkcd, please.

[andrepd]

Feeling anti-joke, are we?

Re:I'd be suspicious

[Anonymous Coward]

You can also go to the modernized pcc. Only 18000 lines to check in 30 files.

This is going to affect the Playstation 4 fork

[DF5JT]

Is it not?

In the long run this is really going to suck for the Sony engineers, maintaining a fork with a different system compiler.

Re:This is going to affect the Playstation 4 fork

[kthreadd]

FreeBSD 10 has been using Clang as the system compiler for a long time already. This was just the removal of gcc, nothing else.

Re:This is going to affect the Playstation 4 fork

[geminidomino]

Doesn't matter, it'll just be yanked out a few months in anyway. ;)

Re:This is going to affect the Playstation 4 fork

[Corngood]

PS4 already uses clang...

Re:This is going to affect the Playstation 4 fork

[tlhIngan]

Is it not?

In the long run this is really going to suck for the Sony engineers, maintaining a fork with a different system compiler.

I don't see why - Sony is not going to track FreeBSD - they already forked it and created their OS around it.

There's no need to track FreeBSD - not that you'd want to because having your dev environment change under you just costs developers money.

Sony's been using BSD for a long while now and unless some major security flaw happens in it that could potentially compromise things, the kernel will probably stay static. Likewise, the compiler too - unless the code it generates has security holes, then it probably will never get updated or even looked at. (And there's still going to be old vulnerable versions out there - Sony can't force everyone to recompile their code).

Even the base OS is unlikely to need recompiling with updated compilers - if it's even bothered with.

The only "suck" happens if newer features of GCC are required in the code. But given the base OS probably won't change unless a major security vulnerability is found, they can switch to LLVM for the application code if newer features are required.

Re:This is going to affect the Playstation 4 fork

[whizzter]

Well looking at the time this changeover has been in progress and that getting the userland and ports into sane order was the latter part of it i would suspect that they might already be running with llvm since the sony userland is probably stripped down and Sony's own code is running most of the show.

That is assuming that the Sony devs had a good clue and sanity enough to track the mother project..

Re:This is going to affect the Playstation 4 fork

[TheRaven64]

You might want to look at the list of LLVM developers and see how many work at Sony some time. In particular, the LLVM Linker (lld) is lead by a Sony engineer. Now take a guess at which product group he works in...

Re:two dictionaries say "no such word"

[rubycodez]

Microsoft will give you a "single pane of glass" for "future stars" to view all the "innovations" and "mentorship" in a "major paradigm shift" to "foster success"

USER? Karma be damned.

[aristotle-dude]

BSD think of the freedom of the distributor, the GPL thinks of the freedom of the user. I see myself as a user, you see yourself as a proprietary distributor.

FFS, I have had it with the "USER" bullshit. Real end users "use" the system and care more about usability than access to the source code. Most of them would not know how to write any code let alone audit it. Stop confusing yourself, an interested third party, with end users. You might also be a user but not all users are interested third parties with coding skills.

The end user's "freedom" is actually limited when nerds argue over licensing terms and are unable to include some software because of it. I am a developer and a user and I find BSD for be free enough. It offers me the choice of proprietary implementation of a BSD project's code, open source BSD implementations and the ability to create my own proprietary implementation for me and my friends without having to hire a lawyer. I can also choose to contribute some or all of my modifications back to the community if I so choose. That is true freedom. The GPL is actually a form of slavery that you agree to enter into in order be able to modify the code. I would rather "pay" money and keep my freedom thank you very much. The price that the GPL imposes is too high a price.

Re:USER? Karma be damned.

[Microlith]

Real end users "use" the system and care more about usability than access to the source code.

Ah, so you get to define who a "Real end user" is I take it?

Most of them would not know how to write any code let alone audit it.

Which is beside the point, really.

The end user's "freedom" is actually limited when nerds argue over licensing terms and are unable to include some software because of it.

Nonsense. You just don't like that the argument doesn't go 100% your way and others disagree.

The GPL is actually a form of slavery that you agree to enter into in order be able to modify the code.

I see we're back to hyperbole. No, you can modify the code all you see fit. You simply can't redistribute without complying with the GPL. Which isn't slavery by a long shot.

The price that the GPL imposes is too high a price.

For you maybe, but you're already acting unreasonably.

Re:USER? Karma be damned.

[Anonymous Coward]

Most of them would not know how to write any code let alone audit it.

Well, yes, and most people don't care about freedom of speech or freedom of information. But that doesn't mean it's not important.

The GPL is actually a form of slavery that you agree to enter into in order be able to modify the code.

Yes, it's slavery: you need to contribute back the changes to your users. But you can use it for any purpose for free.
Man, the developer chooses the license. If you don't like to share, use prorpietary software!

Hey, don't look at me!

[fustakrakich]

I voted for Clodos

Does Clang support all of FreeBSD's platforms

[nogginthenog]

FreeBSD/amd64
FreeBSD/ARM
FreeBSD/i386
FreeBSD/ia64
FreeBSD/MIPS
FreeBSD/pc98
FreeBSD/ppc
FreeBSD/sparc64
FreeBSD/xbox

I guess they dropped 68k...

Re:Does Clang support all of FreeBSD's platforms

[MachineShedFred]

Just like everyone else, everywhere.

Re:Does Clang support all of FreeBSD's platforms

[Anonymous Coward]

FreeBSD never had 68k support AFAIK. NetBSD and OpenBSD do.

Re:Does Clang support all of FreeBSD's platforms

[TheRaven64]

We only made the switch to remove gcc from the default install on amd64, i386 and ARMv6. ARM v6 is still compiled with gcc (clang probably should work okay on ARMv4 and ARMv5, but the people still using them tend to be conservative). It's not yet the default on MIPS, but since we're about to release an open source MIPS core at my day job that runs FreeBSD we're working on improving MIPS support. PC98 is i386. It currently does include gcc, because some oddity in its bootloader requires gcc (not sure what). XBox is also i386. PowerPC 32 is still a little way behind (no PIC support, some issues with TLS), but PowerPC64 is now working fine with clang so will make the switch fine. IA64 is basically dead and Oracle seems to be trying to kill SPARC64 as a platform for anything other than Oracle software, so we'll see if they're still around in a few years. There is a SPARC64 back end in LLVM, but it's not well maintained because no one cares.

Entirely BSD toolchain soon

[Eravnrekaree]

Since the FreeBSD Kernel is BSD licensed, it only makes sense to have the entire base toolchain BSD licensed as well. This is a big step in that direction. There is also a broader project to create BSD licensed versions of all of the core Unix utilities as well. Hopefully we will see this happen soon.

Re:Boycott FreeBSD.

[Eravnrekaree]

FreeBSD is a fully open source operating system, with source code everyone can read and freely share. it is a "Free Software" system. If you mean that it doesnt have a license that tells you that you cannot link FreeBSD code to code under other licenses, then yes you are correct about that, and FreeBSD has always been this way, being a BSD licensed OS, which is why many people choose to use it. FreeBSD already has a BSD licensed kernel and always has.

Re: Enough with the xkcd, please.

[geminidomino]

It's clearly stating some basic facts:

1) XKCD isn't funny.

2) References to XKCD aren't needed here at /., nor anywhere else.

Someone who doesn't know the difference between a "fact" and a "subjective opinion" has no business bitching about anyone else's English comprehension or probability of a learning disability...

There may be some undiagnosed medical condition preventing you from thinking in a clear, rational manner.

Luckily, in you're case, it was identified years ago. I believe the current nomenclature is "recto-cranial inversion." Please, seek the help you need.

Re: Enough with the xkcd, please.

[geminidomino]

And yes, I typoed your/you're. I am now aware of this.

Fuck off.

[Anonymous Coward]

The frigging user is why the fucking software is written: no users, no fucking software you fucking idiot!

Re: Fuck off.

[Anonymous Coward]

wrong.

Re:Fuck off.

[Bengie]

No developers, no software, except in the case where you define the developer as the user, but then GPL says it does not care about the freedom of the developer.. I'm so confused.

Re: Enough with the xkcd, please.

[Anonymous Coward]

1) XKCD isn't funny for me and because I am the president of the universe, nobody has the right to enjoy it.

FTFY and stay mad.

this is really good news

[Lawrence_Bird]

now if we could only move Fotran too.

According to FreeBSD themseves, you can.

[Anonymous Coward]

They say in their attempt to justify not wanting any GPL licensed stuff near their stuff that they believe that free code should have NO STRINGS ATTACHED.

IF this were actually the case (note: it is NOT), then they would be 100% fine with slapping a GPL license on the BSD code.

After all, the ORIGINAL code is still available under BSD, right? That's the line about BSD code being taken propriatory, isn't it? That the original is still there for you to have?

Not, apparently, when you take it GPL.

Why?

Because they have a rabid hate for GPL because they are first, foremosed and only developers and they think users should just fucking sit back and take what they have or be a developer themselves if they want any rights.

Re:According to FreeBSD themseves, you can.

[hedwards]

What are you blathering on about?

They keep the code separate because it's under a different license and by keeping it separate they aren't forced to change the license on all the rest of the code. If you look in the source tree, it's not just GPL code that's kept separately, there's an entire tree for contrib. The code in that module is still GPL code and the resulting binary is GPL when it's compiled in.

Care to cite a reference for where the BSD license permits people to change the license other than the people that own the rights to the code? I'm guessing you can't, because it doesn't exist. Yes, you are permitted to take the code and compile it without providing the source, but the code itself is still BSD licensed code, and nothing is going to change that.

Considering that it's the rabid GPL fanbois that seem to be of the opinion that forcing people to lie about using their code is such a great idea, I'm not sure how much credibility they have. The bottom line is that the code is going to be borrowed either way. At least with the permissive licensing it's likely that they'll post a notice that they're using the code and give back to the community.

The GPL seemed like a good idea at the time, but at this point, this kind of heavy handed licensing is neither necessary nor useful.

Re:According to FreeBSD themseves, you can.

[Anonymous Coward]

They're find with you slapping a GPL license on their BSD code. They just think it's hypocritical.

Re:According to FreeBSD themseves, you can.

[david_thornley]

You can't remove the BSD license without consent of the copyright holders. You can add stuff to it, so it becomes a derivative work, and license that under any license that doesn't have conflicting requirements with the original license. It's perfectly legal to take BSD-licensed code, do stuff to it, and release the changed software under GPLv3, with BSD still applying to the old parts. That's really no different from taking BSD-licensed code, doing stuff to it, and releasing the changed software under a proprietary license, which companies do all the time.

Re:According to FreeBSD themseves, you can.

[DuckDodgers]

This is one of Slashdot's favorite flamewars. :) The Free Software Foundation members believe that all source code should be free, and remain free. It's a very heavy handed license on purpose.

You claim the GPL is neither necessary nor useful. I emphatically disagree. I think Linux would have never become as popular as it is if Linus had licensed it under the BSD license or something similar. Businesses would have made proprietary forks whenever they saw something they wanted, and ignored the core Linux project the rest of the time for fear that anything they contributed back would be used against them by competitors while the competitors kept their own Linux innovations proprietary.

GCC may be stale in some respects now (I honestly have never examined the source, I'm just going on common rumor). But it's enormously common. Would it have received as much attention, investment, and innovation if companies could just make proprietary forks of it? I doubt it.

FreeBSD is an excellent product and LLVM/Clang are excellent too. But I absolutely believe they are only strong because Linux and GCC exist. If Apple wasn't trying to make something open source and equivalent to GCC for their own use, they wouldn't be heavily investing in LLVM/Clang development. Likewise for other companies. In my view, they only have motive to strengthen liberal license open source software because competent GPL software exists. If the GPL software did not exist, or was well outside mainstream use, most big companies would revert to investing in proprietary internal products instead of liberal license open ones. If LLVM is an awesome compiler, that's only because GCC exists.

Re:According to FreeBSD themseves, you can.

[DuckDodgers]

I understand how that would work if I took, say, FreeBSD and LLVM and made a virtual machine, and in the virtual machine I write some piece of code that runs on the FreeBSD and compiles with LLVM, but the code I wrote is itself GPL. Then the release has BSD license restrictions on the operating system and compiler bits, and GPL license restrictions on the code I write.

But if I modify the LLVM compiler, and release the modified version, I can make it part of proprietary software and still comply with the original BSD license. If I modify the LLVM compiler, and want to release the modified version GPL, then I have BSD license code calling into GPL license code, and GPL license code calling into BSD license code. As far as I can tell, that violates the terms of the GPL - all of the code it links with has to be GPL also.

As far as I can tell, that combination can't be done. I honestly don't think the BSD license creators were intentionally trying to sink the GPL with their licensing provisions, but the practical result is that the BSD license is more friendly to re-use in proprietary software than re-use in "copyleft" ( https://www.gnu.org/copyleft/copyleft.html ) software.

So I suppose if you want to make a "copyleft" fork of BSD license software, you would need to use something like the LGPL or the Eclipse Public License. Both of those require all redistribution of the software in source or binary form to include access to the source code of the LGPL or EPL license code and grant the recipient the right to redistribute that code in turn under the same license. But both do not place restrictions on other software that links with or otherwise interoperates with the LGPL or EPL code. But again, I have a very weak understanding of these things.

Re:According to FreeBSD themseves, you can.

[Anonymous Coward]

I'm glad you cleared that up....I was afraid all of the licensing could get stuck in an infinite loop if it were forked and the licensing changed the wrong way....

Re:Boycott FreeBSD.

[Anonymous Coward]

BSD-licensed software is actually more free, if you think about it. The longer the licence, the less freedom it probably has, and is why the BSD community prefer short MIT-style licences. Your ideas about free software have been 'poisoned' by that RMS person.

copyleft (GNU) <--> copy centre (BSD) <--> copyright

are you stupid?

[Anonymous Coward]

Are you stupid?

There's nothing in the post about removing the GPL from the Linux code. Its about removing toe cheese eating RMS and his attempt to rebrand Linux as GNU/Linux from the Linux.

Re:Goodbye statist GPL don't let the door hit ya!

[Anonymous Coward]

yup you're a troll all right.

Linux: TVs, NFS, core routers, Apache, ext3fs, Google.

I think you're full of shit.

GNU is all about philosophical choices

[Anonymous Coward]

You're joking right?

The FSF has been making decisions on philosophical merits instead of technical merits for years; that's why we have Gnome. Its also why their own kernel efforts, such as HURD, keep failing.

And saying: "I'm using HURD right now!" does't cut it, I've hurd them duck fart underwater for the past 15 years.

And so what?

[Anonymous Coward]

If you took one line of microsoft code and put it in your BSD project, you'd be in jail!

And, here's a weird idea: if you want the code, you can ask for permission.

You know, ASK.

Alternatively you can license YOUR code as GPL. That's free according to the FSF too, which you appeal to as authority on freeness.

And guess what: that code you made GPL that was yours: YOU CAN STILL RELEASE IT BSD! Just not with that one line of GPL code.

If you don't like this, then get copyrights ruled null and void.

Re:And so what?

[Bengie]

And, here's a weird idea: if you want the code, you can ask for permission. You know, ASK.

Yes, dig through GIT and track down the 20,000 people who may have touched those 5 lines of code and get written permission. Sounds great!

143,409% wrong.

[Anonymous Coward]

No, the GPL is not about making you open source your code any more than the BSD is about making you give your code away for nothing.

Your raving sounds like insanity to me.

Re:143,409% wrong.

[Bengie]

Because you can make changes to GPL'd code and distribute the binary without the code you used to compile it with?

Kind of like saying the drunk driver didn't kill that person, excessive g-forces did. Just a technicality with no practical merit.

I'm blathering on about the FreeBSD statements.

[Anonymous Coward]

Haven't you read them?

They state in black and white that *truly free* software should have no strings attached.

Absolute verbatim quote.

If "You cannot slap a GPL license on this code" is not allowed, THEN THAT IS A STRING ATTACHED.

Was that too complex for you?

Re:I'm blathering on about the FreeBSD statements.

[hedwards]

That applies in all cases. Even with public domain software. You cannot add a license unless you own the copyright to it.

Re: Enough with the xkcd, please.

[Archangel Michael]

Cranial Rectomitus is the term for this affliction that I learned. Has it been updated?

Re:Boycott FreeBSD.

[Anonymous Coward]

Boycott? I'll compromise with you on that. I'm going to keep using it, but I'll refuse to pay for it.

Not a Big Deal.

[srobert]

In practice, if you want to install a port on FreeBSD that needs to be compiled with gcc instead of clang, the installation will just pull gcc in as a build dependency. Other than the fact that the compilation will take a little while longer, you probably won't even notice it.

Re:Boycott FreeBSD.

[Anonymous Coward]

It's not free software. Only GPLv3-licensed apps are truly free. You're talking out of your ass.

CLANG doesn't support 68k.

[Anonymous Coward]

So, fail there.

Which hardware that FreeBSD supports that GCC doesn't (and please include what FreeBSD supported before they decided to move to CLANG because CLANG didn't support it, because if that's an acceptable change, then its just as acceptable to drop whatever hardware you think GCC is no longer supporting)?

Re: Enough with the xkcd, please.

[Anonymous Coward]

Fuck off, you stupid troll, and DIE.

8,346,892% wrong.

[Anonymous Coward]

TdR went APESHIT when someone put code that they wrote that had been dual licensed in an updated version but decided to make it ONLY GPL.

TdR went APESHIT when someone took some code that was dual licensed and removed one of the licenses.

THEY ARE NOT FINE with you slapping a GPL license on their BSD code. They're fine with it being closed completely. That is because they're hypocrites.

Re:8,346,892% wrong.

[Anonymous Coward]

Only the copyright holder can change the license. Theo was 100% in the right.

Re:8,346,892% wrong.

[smash]

removed one of the licenses.

I've just decided I'm going to remove the license for the software I have on my machine because I don't like it. I'm sure Microsoft and Apple will be just fine with this, as the precedent has been set by somebody doing this with BSD code and replacing it with GPL.

for the sarcasm impaired: you can't REMOVE somebody's license simply because you don't like it.

Re:8,346,892% wrong.

[1s44c]

TdR is the Gordon Ramsey of UNIX. Mad Skilled but he goes APESHIT whenever he sees something he doesn't approve of.

But if you take code and change the authors license that's stealing. Just because you don't like their license to begin with doesn't make it OK. Unless the license was written in such a way that it was permitted to permanently relicense the code and derivative works Theo was in the right here.

Not all BSD developers

[Anonymous Coward]

THEY ARE NOT FINE with you slapping a GPL license on their BSD code. They're fine with it being closed completely. That is because they're hypocrites.

I think you mean Theo de Raadt is not fine with the GPL being added to BSD code. He's not the only one for sure but I don't think THEY (*all* BSD developers) are against it.

What does this mean for OSX and Xcode?

[Monsuco]

I don't really care about the philosophical implications. What I want to know is if I'll still need to download Xcode from the App Store to get a copy of GCC.

Re:What does this mean for OSX and Xcode?

[Anonymous Coward]

This is completely orthogonal Mac OS X.

Re:What does this mean for OSX and Xcode?

[Guy Harris]

What does this mean for OSX and Xcode?

Nothing. The subject line says "FreeBSD", not "OS X" or "Xcode".

What I want to know is if I'll still need to download Xcode from the App Store to get a copy of GCC.

Then ask Apple whether 1) the only way to get a compiler from Apple will continue to be to download Xcode and 2) whether Apple will continue to provide GCC.

You forgot...

[Anonymous Coward]

It's also why gcc and glibc are unmaintainable piles of shit.

Go ask people who were around back in the early days. In order to avoid people building closed source frontends and such on top of gcc they mucked around with the interfaces such that few people could understand how they worked, nevermind keep up with the internal changes.

Not that I can blame them given the era, but that cruft was supposedly a lot of the mess either leading up to the egcs releases or leading up to the gcc 3 series (which was also mucked up by redhat creating 'gcc-2.96' because they had to have 'standards compliant c++' before everybody else, and in the process lose backwards AND forwards binary compatibility across almost the entire toolchain/system.)

Moreover though...

[Anonymous Coward]

I thought llvm only ran on i686 hardware, in which case you couldn't have an i386 self hosting toolchain for FreeBSD anyways (Interesting to consider given that NEITHER clang nor gcc would be capable of compiling anything on an i386 to begin with simply because any non-trivial program will take up more RAM than any 386 and possibly most 486s were capable of addressing.

Re:two dictionaries say "no such word"

[Guy Harris]

Microsoft will give you a "single pane of glass" for "future stars" to view all the "innovations" and "mentorship" in a "major paradigm shift" to "foster success"

But only if you proactively leverage your synergies.

Re:two dictionaries say "no such word"

[Bengie]

Trolls and anti-trolls have a symbiotic relationship. Two faces of the same coin.

Re:I'd be suspicious

[Anonymous Coward]

The change is a year old; it sort of made sense at that time.

Re:I'd be suspicious

[TheRaven64]

RDRAND is not used directly, it's 'whitened' by Yarrow or a similar algorithm (there are now a few pluggable options, although I'm not sure they made it into 10.0). It's one entropy source, and things like interrupt timings and so on are others. If RDRAND is produces non-deterministic data in a way that are easy for the NSA to predict then there are some possible network attacks involving TCP sequence numbers early on in the boot. After that, there are other entropy sources (e.g. interrupt timings) that make it much harder to predict the inputs to Yarrow (and, unless Yarrow is also broken, you need all of the inputs to be able to predict the output).

Re:Boycott FreeBSD.

[TheRaven64]

Seriously, the core devs hate free software

Speaking as a member of the FreeBSD Core Team, who contributes code to a GNU project (GNUstep) and has signed an FSF copyright assignment for the purpose, and who was the one to make the disable-gcc-by-default commit, I have to say: What?

Re:who's heard of phoronix, and GPL3 vs all comers

[Anonymous Coward]

The 'phoronix' benchmark and phoronix website (add dotcom to see for yourself) of which you speak were also new to me until last year or so. I have no financial interest in them, just a technical interest -- I like fast libre laptops, and phoronix gives me the hard data on how to build them. They benchmark straight from git, in many cases. Basically, the idea of the site (and the motivation for creation of the benchmarking suite) is something like what Tom's Hardware and Anandsoft do, i.e. performance review of PC hardware especially frames-per-second numbers, but with the focus entirely on the software stack on Linux -- plus some *BSD articles for spice -- with particular emphasis on the graphics stack, in other words, the 3D gaming speed of Linux. The benchmark suite from phoronix is actually a meta-benchmark: it contains 700 kajillion benchmarking tools, from the most inane microbenchmark up through an entire copy of a 3D game, in a nice packaging system with a TUI. You can bench gcc versus llvm in 50 different ways with phoronix, in other words. Like rotten tomatoes film reviews (where they collect a hundred individual reviews of LOTR and then give it an overall score based on how many reviews were glowing versus how many reviews talked trash) the phoronix meta-benchmark of system performance is basically impossible to game.

As to the particular testing of LLVM, by phoronix, a little background may help. One aspect of 3D gaming is OpenGL, and with both wayland and mir depending on OpenGL ES under the covers -- not to mention android -- phoronix is busy testing the various aspects of OpenGL ES performance. One of those aspects is shader performance, i.e. the little sub-programs that OpenGL apps like games and CAD systems generate which run fully inside the GPU hardware, and some versions of the X.org graphics stack have started to use LLVM to jit this shader-code on the fly. There are also some generic-system-performance tests that phoronix would probably use, to exercise the system-compiler. The site spends a lot of time doing graphics-stack testing, like comparing libre GPU drivers versus binary-blob GPU drivers on ten different graphics cards and such, but it also provides distro comparisons that rate overall performance of the new distro compared to the competitive siblings (and to previous distro-generations).

So, the answer to your question is, prolly clang devs do not care about phoronix. They should, but mostly what motivates them is their paycheck from apple, which as comments above have explained, is the driving reason for BSD junking gcc prematurely -- apple cannot use GPL3 inside their walled iOS garden, or prolly their walled OSX garden for that matter, and thus gcc had to go. (Yes -- I realize that most FreeBSD contributors are not working for apple. But that is not the case for clang -- most of the devs there are apple-funded.) There are many use-cases where gcc is still faster. One way to overcome that is to 'force' the fully-baked but not yet fully-icing-ed (is that even valid grammer?) clang onto the BSD endusers. There are enough people that will use FreeBSD v10, that apple will get a nice beta-test of how well clang works as the sole system compiler, so they can pay the clang devs to work on the deficiencies prior to the next OSX upgrade. Apple changed over to using llvm and clang as the default a couple of years ago in osx 10.7, see here -- http://stackoverflow.com/questions/6796982/clang-and-the-default-compiler-in-os-x-lion -- but the FreeBSD userbase tends to be *way* more technically inclined than the average person in the OSX userbase. In a nutshell, apple got tired of testing for llvm and clang performance regressions and the like all by themselves, so they are getting the typically-more-bleeding-edge FreeBSD userbase to do this for them, starting with FreeBSD v10.

(end of part one, who's heard of phoronix, beginning of part two, GPL3 versus all comers)

Now, I'm pretty grumpy about the whole situation (if there is anything that shows why GPL3 is

Re:fuzzy on licensing

[Anonymous Coward]

You are 100% correct that anybody can take freebsd and make a proprietary product (e.g. OSX, and in a previous generation, Solaris/Ultrix/etc), and that anybody could take clang/llvm and make a proprietary fork (though in practice methinks ibm/apple/etc will focus on making proprietary plugins for it instead). hedwards is just wrong in saying you cannot *effectively* relicense -- he is being pedantic and saying 'technically' you do not really relicense the original you just release something nigh-identical which it just so happens is under new license and/or binary only. This distinction, copyleft-cannot-relicense, and permissive-can-effectively-relicense, is *the* fundamental difference between the GPL (or any copyleft license) and BSD (or any permissive license e.g. MIT license or for that matter public domain). If the existing codebase is under a permissive license, you can fork it into a proprietary product, or even into a copyleft project, including GPL3 -- one particular example of the latter is the drive-cloning software G4U, which some whippersnapper forked into G4L, converting from a BSD license to a GPL license in the process, and making the G4U author unhappy -- but there was nothing he could do.

However, the bare fact that llvm/clang is under a permissive license does *not* mean your suggested strategy, that the FSF can merely bide their time, and eventually take the llvm/clang codebase and produce a GPL3-encumbered fork, actually makes any sense. The reasons are subtle, but I'll try to keep them short and sweet. First of all, it is likely that apple and ibm will treat the freebsd version of llvm/clang as a reference-implementation of the compiler, and in practice will ship 'tweaked' binary-only versions with OSX and whatever operating systems IBM uses nowadays. That means that your hypothesized llvm-derived GCCv9 (to give it a name for this discussion) under GPL3 would actually be slower on OSX than the AppleBinaryLLVM shipped with the same system, because only Apple's would have the key tweaks. hedward would respond like this: "At least with the permissive licensing it's **likely** that they'll post a notice that they're using the code and [**likely**] give back to the community." emphasis added by moi. In practice the posting notice happens -- BSD license requires it ... but whether the second part happens or not depends ... and mostly what it depends on is how big the company in question is, and whether they can afford to maintain a private fork. In practice this happens regularly: OpenJDK is open source, and even GPL, and both oracle and google contribute patches back to the code... but OracleJ2SE for windows has proprietary tweaks not found in the Linux OpenJDK, and because oracle owns the copyright, they can "relicense" the JDK (to themselves) and release binaries for windows. Google does not distribute their private fork externally, since they don't have the copyright (Oracle does); but they use their private OpenJDK fork for their thousands of company-internal servers. Whether either group contributes back is mostly a question of convenience, and the bottom line.

Second, clang/llvm is explicitly designed to permit proprietary plugins slash extensions, so even if we assume there will not be any private tweaks in private branches, certainly the FSF won't be able to fork the proprietary *extensions* of the clang/llvm codebase. Most of the valuable bits will reside in proprietary extensions, one might predict. For an example of this phenomenon, see the magento shopping-cart-app under the OSL: code for the core is freely available, but without proprietary extensions, many store-owners cannot get their system to function well enough for their needs. Contributions back to the core are minimal; everybody concentrates on building proprietary plugins, not on solving issues in the core (key example, relevant to the clang/llvm versus gcc question -- stock magento is horridly slow, because proprietary performance-plugins are the bread and butter

Re: spirit of the FSF

[Anonymous Coward]

There's only *some* spirit of the FSF in Linus himself, who prefers GPL2 over GPL3, but that is not the same thing. There's only *some* spirit of the FSF in RedHat Linux distros (Fedora / RHEL / CentOS / SciLin / OraLin), because portions of their code are BSD/MIT/X/similar license, including their new Wayland project... but large portions are GPL2 or GPL3. Ditto for Canonical's Ubuntu Linux distros (LTS/nonLTS/kubuntu/xubuntu/lubuntu/crunchbang/puppyLTS/etc), which has GPL3 for Mir, albeit with the CLA: they have *some* spirit of the FSF. Every Linux distro has some code written by FSF employees; the question is how much. Every Linux distro has some code under FSF-designed licenses; the question is how much. Every Linux distro has code written by RMS personally; the question is how much. Maybe it will help if you break down the meaning of the "spirit of the FSF". That can mean:

1. use of GPL2/GPL3/LGPL2/LGPL3/GFDL licensing, or even of other FSF-approved but not FSF-originated licenses
2. use of GPL3/LGPL3 licensing specifically, the licenses the FSF is most approving of and second-most-approving-of
3. use of projects hosted at gnu.org -- emacs, gcc, binutils, and a slew of others
4. use of code and/or philosophical positions of RMS personally, as founder of FSF and GNU
5. some combination of the above

Unless you specify which of the numbers you are talking about (and what exactly you mean by 'Linux' ... just the kernel? just Linus Torvalds?), it makes zero sense to say there is no spirit of the FSF in Linux, or that they are were never associated in any way. Those statements are so badly phrased they aren't even wrong, as wolfgang might say.

Re: true spirit of freedom

[Anonymous Coward]

You gave a very solid explanation of both sides of the old debate: picking GPL for your code means that code is henceforth libre forever (available to you and many future folks -- as long as they are *also* the sort of folks that want *their* contribution to your code to similarly become libre forever). Alternatively, picking BSD for your code means that code is libre immediately, to you and to everybody else, even if they disagree with you about libre licensing (the consequence being that *some* versions of the codebase, variants that are not the one you originally released as BSD, *might* no longer be libre at some point, or might get converted into GPL at some point, or whatever).

I would argue which system is "truly free" aka "really free" depends on what you mean. If you mean, this code and any derivatives thereof, will be libre *forever* to myself and to others that respect forever-libre, then the GPL is really & truly free-as-in-freedom. Alternatively, if you mean this *immediate* version of my code will be libre forever, but derivatives thereof might be libre-immediate-bsd, libre-forever-gpl, or libre-no-longer-proprietary, then BSD is really & truly free-to-everybody-regardless-of-politics.

It *is* good both choices exist. Rather than saying 'really' free and 'truly' free, though, which are biased terms that do not convey what you specifically mean by the adjective, I would suggest distinguishing between the BSD as being this-version-of-the-code-libre-forever but derivatives of the codebase can undergo relicensing (by anybody including proprietary folks), and GPL as being this-version-libre-forever, and all derivatives also guaranteed to be libre-forever (which not every future contributor will be willing to accept). That sort of phrasing is reasonably neutral, and highlights the advantages of both licensing options. As you say, which option you pick, as an author of some new project, depends on what you want. Linus w/ GPL2, RMS w/ GPL3, Bill Joy w/ BSD, and Tim Berners-Lee w/ public-domain are good examples here of reasonably famous picks.

BSD has the advantage that future derivatives of the code can become closed source (or be converted to GPL for that matter). GPL has the advantage that future derivatives will always be GPL. Subject to people obeying copyright and contract law, of course.... "Just don't ever think that because your code is GPL'd, no one will dare to use it in close source." This is a good point. The GPL, especially GPL3, definitely depends on enforcement via lawsuits, grounded in the legal term 'propagate'. If you suspect that some competitor is using your GPL3 code, hidden behind a jailed bootloader, or locked up in the cloud, the only way you can stop them is with the threat of legal action. That's a big stick, of course, and with subpoena power you could in fact find out whether they were swiping your GPL3 code or not. But the whole approach relies on lawyers; distasteful. BSD also relies on lawyers, but only for the older four-clause licenses that *required* notice, and most BSD people would not see omission of notice as such a big deal, whereas most GPL people would see omission of GPL'd source code for any propagated derivatives as a huge deal indeed.

(p.s. as for apple 'stealing' stuff, that is also a loaded term. Pretty clearly andreicristianpetcu was upset about exactly that, given his reference to "pretty iShit" at the end... but also look at what he said first, which is not that Apple stole something from him, personally, but that Apple took what was once communal property available to everybody, and made a derivative that was no longer communal property. Which is, basically, complaining about the core feature of the BSD license itself -- such events are guaranteed to happen, sooner or later, with any successful BSD-licensed project. Something very similar, a hardware company commercializing some software that was formerly available to everybody, and thereby impacting the community negatively, is exactly what prompted RMS to in

Re: true spirit of freedom

[santax]

Thank you for your lengthy response to what I said. I have to agree with you. I used the words true/free based upon my own bias and I shouldn't do that. Good point, I will take that into account for the future. Especially since it's such a 'holy' (as in vi vs emacs) subject. About the ps: I personally don't see it as stealing what Apple did. I consider the BSD license as sort of saying: Here, you can have this. It's a gift. Do what you want with it. And albeit not being too keen on Apple for other reasons, they had every right to use the codebase under the BSD. So no stealing as far as I am concerned. It's funny you mention RMS. I have great respect for him and consider him to be one of the pioneers of the tech-world. When he speaks, I listen. I don't always agree with him 100%, but most of the time I think he's right. To be honest, when I started reading your post for a moment there I thought I got AC'd by him. Again, thank you for this very interesting response. I wasn't expecting such a post from an AC.

Re: endusers vs programmers&distributors

[Anonymous Coward]

Aristotle-dude is trying to make an argument from the vox populi ... ironic, considering what the actual Aristotle had to say about the mob. The point that most people running OSX/Windows/whatever are not programmers is related to his point about "real" endusers. Think of it in terms of the so-called Average Merrikan: they don't know how their computer works, they don't know the difference between GPL and BSD, all they care about is that this black box they bought has good value, which is defined as roughly having plenty of features/stability/speed/style/whatnot for the price. They could care less if the black box was produced by sacrificing oompa-loompa children to the fourth level of the underworld, or about the kind of text-editor which was used to write the software, is the gist of it.

His point is, rather than argue about whether to use GPL or BSD, i.e. argue about morality of licenses, and the *future* state of the software industry, programmers should instead just concentrate on building the lowest-cost highest-featured black boxes possible. If that means spilling some oompa-loompa blood, and risking unleashing the evil residents of the fourth level to feed upon the helpless humanoids on this plane of existence, and editing with ed (The Standard Text Editor), the endusers will not care. Well... until they are gobbled up by some multi-headed troglodyte straight out of a graphic novel, that is.

The trouble with this argument is twofold. First, it assumes that there is no connection between how software is produced, and how software is licensed, and the ultimate long-run happiness of the endusers. This is clearly wrong: the sort of product, and the sort of price, that is produced by a GPL/BSD/proprietary licensing model definitely differs, and more importantly, over time these approaches produce dramatically different economic outcomes for the enduser. (Just because the enduser is unaware of these impacts, or at least, of the underlying causes behind these impacts, is not relevant.)

Second, it assumes that morality stops at the edge of the black box. Notice how he accuses the GPL of being a form of slavery (on programmers), while simultaneously arguing that endusers (implicitly defined as exclusively non-programmers) don't care how software is produced. Reminds me of arguments about the politics of globalization -- people will argue that non-union payscales are slave wages, and then turn around and complain that endusers don't care how goods are manufactured. The correct result for software is a combination of moral licensing -- which usually means BSD or GPL depending on what specific flavor of morality you want to uphold -- and community standards, which is inextricably intertwined with the choice of license, since in the long run what matters is how they treat derivative works. As for globalization, methinks the correct result for economics is a combination of capitalist economics coupled with self-evident individual rights... most countries get one or the other wrong nowadays... but that is another story for another time.

Anyways, the key sentences from his diatribe are these: "I find BSD for be free *enough* [for endusers whether programmers or not]... price that the GPL imposes [on programmers&distributors] is *too* high..." emphasis added. Ignoring the hyperbole, he clearly understands that we're talking about matters of degree here. But small differences, which accumulate over time, can be magnified into much larger differences that impact the systemic nature of the way software is produced. That is true of economic systems, as well as of software licensing.

Re:fuzzy on licensing

[DuckDodgers]

Thanks for the thorough pieces of information. I learned a few things, and with the stuff I didn't learn I agree on most. However, I'm going to disagree with your last paragraph.

I think a fundamental reason most modern websites are written in languages like Perl, PHP, Python, Ruby, or even Java is that those languages have a much faster compilation cycle than C or C++ for complex applications. Of course the scripting languages are compile-and-run in one fast step. For certain classes of applications where speed is essential, you can't escape the need for C or C++ or something similar. Otherwise, I think the speed of the change-save-run cycle trumps the speed of the execution in importance by a wide margin.

A company with a big war chest of money like Google, Microsoft, Oracle, or IBM can pick (or build) the compiler that lets their 100 servers serve the results as quickly as possible because they're battling costs associated with thousands of servers. But for some company that's looking at 50 servers, or 20, or 3, the company with the quicker compiler probably has his product on sale while the other company is twiddling its thumbs. Whenever I've had to stop what I'm doing to wait for something to finish at work, inevitably I switch from being productive to wasting my time doing something else (like, just for the sake of argument, posting to Slashdot).