Slashdot Mirror
FreeBSD 10 To Use Clang Compiler, Deprecate GCC
An anonymous reader writes "Shared in last quarter's FreeBSD status report are developer plans to have LLVM/Clang become the default compiler and to deprecate GCC. Clang can now build most packages and suit well for their BSD needs. They also plan to have a full BSD-licensed C++11 stack in FreeBSD 10."
Says the article, too: "Some vendors have also been playing around with the idea of using Clang to build the Linux kernel (it's possible to do with certain kernel configurations, patches, and other headaches)."
What's wrong with GCC?
and suit well for their BSD needs
In other words, it has the license they want. GCC was a thorn in the side of the collective BSD conscience, they didn't like it.
"First they came for the slanderers and i said nothing."
Having all this great open source compiler technology competing with each other is great, but one does wonder if the alienation caused by GPLv3 was worth it, as it is the primary reason both Apple and FreeBSD embraced Clang (in fact, Apple started the Clang project). As a result, GCC wasn't updated past GPLv2 on either platform. Apple couldn't integrate GCC with their IDE like they wanted, nor could FreeBSD's commercial clients work with it. Flexibility and pragmatism usually wins out over rigidness and ideology.
I've heard positive and negative claims regarding this. Certainly, Apple thinks it's production-ready (I think it was Xcode 4.2 that they stopped shipping GCC). Do you have a link showing that generated code is significantly worse? Which versions were compared?
What's wrong with GCC?
Some people argue that LLVM/Clang offers better code generation, compile time warnings, and code analysis. Some compiler developers think the gcc code has become too bloated and complicated. Even gcc devs have described the gcc code as "cumbersome".
There are various efforts to get Linux building under LLVM/Clang. Especially for embedded environments.
Stallman and others deliberately fought having APIs, proper documentation and to allow plugins for all parts of the GCC toolchain, to keep control of the thing.
Mostly result of dispute with DEC SRC when GCC and parent FSF failed to enforce GPL on Modula-3. Moving target known as GCC internals has been problem ever since, mostly to "legitimate" GNU compiler developers.
LLVM, on the other hand, made ingenious move with standard and open IR. Overall modular design is another boon.
GCC was in blind alleys before. No real reason for them not to survive this one. Another EGC can happen, to pull GCC in future.
http://opencm3.net, http://www.nongnu.org/gm2/
I've found that code that will compile properly under a variety of compilers tends to be of better quality.
One of my current projects started out on an old 2.x branch of GCC. When I finally got around to updating to a current GCC, I had to fix quite a few bugs before it would actually work - the different compiler was catching problems I hadn't noticed before.
Same when I tried compiling it under Visual Studio, or Clang - the more compilers I made it work under, the less bugs there were in the code.
Now, if a given program actually uses some special feature of GCC, that's fine - if only one compiler will do what you need, that's fine. Or if it's too much work to maintain a "port" - I stopped maintaining the VS project files a while ago, since I no longer used it. But if you have a chance to at least test it against a different compiler, go ahead and give it a shot.
But Clang also has many technical superiorities to GCC too. Wikipedia gives a quick outline of them.
http://en.wikipedia.org/wiki/Clang
And at the top right of that page is this picture. Which Pokemon is that?
... Which is hilarious because it is the BSD fundamentalists who are re-implementing huge projects just to avoid a license they don't like for no reason other than political correctness ...
Untrue. Gcc is handicapped by political decisions in it technical design. It intentionally does not allow "others" to plug into some "internals". "Internals" that would facilitate other tool builders, especially those creating a graphical integrated development environment.
LLVM/Clang doesn't come with such technical baggage. Its modular rather than monolithic. It is a newer code base that is far easier to work with, even gcc devs moan about the bloat/complexity of their code base. Nearly all long lived project reach a point where it is better to toss the legacy code out and start from scratch, gcc may have very well surpassed that point.
And on some platforms LLVM/Clang simply generates better code.
The quality of the generated code, however, is significantly worse, at least at this time.
That may not be accurate. My understanding is that on some platforms LLVM/Clang has the advantage.
Complaining about the GPL is like complaining that you can't play dirty pool with code licensing(see Tivoization).
I haven't heard Apple complaining about the GPL or trying to circumvent it - they're just switching to alternative projects.
Of course, its a pity, because even if if you Tivoized GPLv2 code you still had to share your source so people could learn from it, or use and modify it on other (or jailbroken) hardware, whereas now people are moving to BSD-style licenses with no such benefits... but if the FSF want to let the perfect be the enemy of the good, declare jihad on Tivoization and have a tilt at the patent windmill, that is their right.
In a survey of 100 programmers, 111111 thought that duck-typing was a good idea.
There's a reason even the shining monument of GPL (Linux) uses GPLv2...
Even if Linus did want to move the kernel to GPLv3 (he doesn't), he would have to get every kernel contributor to agree to the license change, AFAIK.
Do what thou wilt shall be the whole of the Law
Complaining about the GPL
He wasn't complaining about the GPL, he was stating (correctly) that it is one of the key reasons some groups choose not to use GCC, particularly GPL version 3 (note that FreeBSD has not used and of the recent GCC release specifically for that reason - they were fine using GPLv2).
It is a case of choosing the right tool for the job. Until recently their choice was to redefine the job (change the parts of their projects and licensing policy that GPLv3 conflicted with), keep the old tool (using older, GPLv2 licensed, releases of GCC only), or use something less stable/proven/compatible. They chose the middle option. Now option three is replaced by "use something else that is now stable/proven/compatible enough to be an alternative" they have taken that choice. Again this isn't complaining about GPLv3, it is simply refusing to use it because it is incompatible with some of their chosen goals.
I'm told there are technical reasons why Clang and the related tool chain are preferable to GCC in some circumstances too, though I'm out of the loop on that one so I don't know what they are or if they are significant to FreeBSD.
While his answer was rather terse and it would have helped to be less so (it made him appear to some, to you at least, like an anti-GPL troll), what you appear to have done in your response is set up a strawman to attack. This is the sort of thing that anti-GPL people (both within the open-source arena and external to it) will jump on as "proof" that GPL advocates are rabid loonies, so by defending the GPL in such a manner you may be harming the cause rather then helping it. You might want to be more careful not to come across that way (it may not have been you intention on this occasion, but it does seem that way by my reading).
Many people wants _real_ freedom for their software
No, they don't. If they did, they'd use GNU GPL v3.
But I think you mean the opposite of what you said: people want to be free to do whatever they want with the software, including taking away the software's freedom.
That's the thing. Free software is not about your freedom, it's about the software's freedom. It is not for the benefit of anyone in particular, it is for the benefit of the whole humanity. When you think about where rms came from, and when you read his writings, you realize that his ideal is not an indifferent "here's some code, use as you wish". It is an ideologically grounded "here's some code, it's for everyone to use, and if you build upon it, the result is also for everyone to use".
Circumcision is child abuse.
I think one of the heads of Red Hat nailed it when he was asked about RMS: "Richard treats his friends as his enemies". Whether the community wants to accept it or not when RMS specifically targeted a single company with GPL V3 he gave a pretty damned good reason for businesses to stay away from the GPL, fear of being the target of GPL V4. BTW I personally bet that if there is a GPL V4 the new buzzword will be "Androidization" since RMS hates Android even though it has put more Linux devices into users hands than anyone else in history.
What I do find interesting though is how many are having a shitfit when more and more move away from GPLed software when....isn't that the whole damned point of the FOSS model? that one man or one company can't dictate to the masses and the community can route around damage? Well that sure looks like what is going on to me, RMS made GPL V3 too nasty and won't come to the table and work with devs on a compromise so they are routing around the damage by choosing different licenses. Personally I figured the community would be happy as it shows that the FOSS model can't be broken or controlled by any one man and with so many licenses out there...BSD, Apache, MPL, AGPL,etc nobody has to "take it or leave it" like they do with the big two.
ACs don't waste your time replying, your posts are never seen by me.
In a couple of years time, there will be a proliferation of different, incompatible versions of CLang/LLVM that will be increasingly expensive to maintain. Furthermore, I can foresee vendors making incompatible changes to the code produced by CLang, subtle ABI breakage and the like. The upper levels will suffer too : vendor A's version will not be able to compile source code with vendor B's extensions and vice versa.
Hindsight is invariably more accurate than foresight. And in this case, hindsight tells us that there are plenty of non-GPL free packages that you use every day that haven't succumbed to either of your fears. In fact you use at least a couple of them when you read this.
This sounds like the 1980s/ealy 1990s all over again
That wouldn't be bad. The productivity per user has never been higher, and most of what we use now was invented then. I'd rather see that again that these modern days where ideas are scarce and productivity per user base at an all time low.
Of course, its a pity, because even if if you Tivoized GPLv2 code you still had to share your source so people could learn from it, or use and modify it on other (or jailbroken) hardware, whereas now people are moving to BSD-style licenses with no such benefits... but if the FSF want to let the perfect be the enemy of the good, declare jihad on Tivoization and have a tilt at the patent windmill, that is their right.
This is absolutely the case! When TiVo was complying w/ GPLv2, the FSF suddenly discovered a major objection to their practice - namely, that they were putting the code in read-only devices, and declared a jihad on the company. However, even GPLv3 doesn't explicitly say that GPL software cannot be put on a Read-only memory (which would again violate the GNU's Freedom #3) or copy-protect memory (which could prevent the device that contains the software from getting copied) or anything else about the devices that the software can reside on.
As you very well put it, it's one more of those cases of the perfect being the enemy of the good, and in the process, the FSF waging a war on its own licensees, namely TiVo. Given that track record, which company in its right mind, even if they endorsed the liberation of software, would want to get into bed w/ the FSF?
Here is what I personally don't get, maybe someone can explain it to me, but WTF was it with RMS and the TiVo? It was ONE device that had NO choice but to be made the way it was. It wasn't like TiVo was being run by Cobra Commander here, they knew that if there was a way to get the content off the device (which is EXACTLY what running custom versions would have allowed) it would have been banhammered in the west quicker than you can say "copyright infringement".
Would he have been more happy if it had used WinCE? Because that is what it feels like to me, that RMS wants ONLY true followers of "the way' to use anything GPL. And since there is multiple other OSes out there including BSD and Windows Embedded it just seems stupid to attack one specific corp and make other businesses afraid of being next on RMS' shit list and all over a device that frankly could have been made no other way without only being sold at China Mart and other "pro piracy" hardware sites.
ACs don't waste your time replying, your posts are never seen by me.
Static for functions or variables with file level linkage makes them "private" to that file. E.g. in this case, several source files can define global variables named world_type without collisions. That is provided all declare them static. One of the files might ommit, but if two or more source files declare non-static global variables named world_type then the linker will (correctly) complain when linking.
When you are sure of something, you probably are wrong (search for "Unskilled and Unaware of It").
In a couple of years time, there will be a proliferation of different, incompatible versions of CLang/LLVM that will be increasingly expensive to maintain.
It's already happened. This is why so many companies are now actively involved in the LLVM community: it's cheaper. I'm currently on my way back from BSDCan (where I was talking a bit about the progress in switching to clang) and I was at EuroLLVM a couple of weeks earlier. Both conferences were full of corporate contributors to LLVM and FreeBSD (two projects that I work on). They like the fact that the license means that they don't need to run everything that they possibly want to do past their legal team and, over the past decade, they've all discovered (at different speeds) that it's much cheaper to engage the community and push work upstream than it is to maintain a private fork.
You get much better support from companies that join your community because they regard it as being good for them than if they dump code on you because they are legally obliged to. We don't want drive-by code dumps, we want long-term commitments to maintenance.
I am TheRaven on Soylent News
According to Kernighan and Ritchie, the static modifier restricts the scope of externally declared variables to the rest of the source file. AC might not want to accept the GPL and BSD definitions of free/unecumbered to non-software contexts.
I am TheRaven on Soylent News
GCC doesn't need GPL, a freeware license would suffice.
Over the years I've read and analyzed the sources and consistently found the source to be as readable as the binary code, regardless of platform.
Compiler developer forums seem to confirm my findings with frequent comments and insults on GCC like:
Nonetheless, I praise the omnipotence of GCC, as long as I'm not the first guy on a new platform.
Tivo did exactly what RMS started the Free Software Foundation to prevent (The Printer Story). What did you expect would happen?
Here is what I personally don't get, maybe someone can explain it to me, but WTF was it with RMS and the TiVo? It was ONE device that had NO choice but to be made the way it was.
If it had not been TiVo it would have been something else. His problem was the use of GPLed software in that manner. While it wasn't found to break the letter of the license it broke the clearly stated spirit of the license, so that wording was updated in v3 to patch the hole.
Would he have been more happy if it had used WinCE?
Yes, basically. Or some form of BSD (the licenses used there would allow this sort of use IIRC). Or anything else not GPL licensed. They had those choices available to them.
RMS is an absolutist on this and similar matters (some would say extremist, but I feel that label to be rather too strong here): if you want to use Free, keep it Free with your use, otherwise use something else (paying for it if need be).
it just seems stupid to attack one specific corp
He wasn't going after one specific corp, just the first one that did it (visibly) first and shoring up the hole before others tried. Remember that TiVo could keep using GPLv2 software as they had already done, they'd just have to start maintaining by other means once later versions switched to GPLv3, so the switch to GPLv3 did not explicitly stop them distributing their product.
and make other businesses afraid of being next on RMS' shit list and all over a device that frankly could have been made no other way without only being sold at China Mart and other "pro piracy" hardware sites.
That is where it falls down of course, but so does every other license commercial or otherwise - if you can't enforce the license in a territory people wanting to do something against the (letter of the) licence in that territory are at an advantage to those elsewhere. "Pro piracy" regimes are not a GPL specific problem and not really relevant here - you could just as easily state that VMWare's recent licensing model changes are an attack on compliant companies.
There are many people who think RMS is wrong on the matter, of course. Linus for instance still explicitly uses GPLv2 as evidenced by it being the license git is released under (the kernel is a different matter: that could not be switched even if he wanted because of how many contributions there have been where rights were not explicitly handed over to the project).
I'm a 10 year+ FreeBSD contributor. You're all missing the point. Linux and BSD target different markets and are optimized in all ways, organization, release process, license, code, to fit these different needs. One isn't better or worse. Obviously Linux is larger in all ways than BSD but larger doesn't mean better or we'd all just be using windows. This isn't a question of llvm being better than gcc, bsd being better than linux, or bsd license being better than gpl. They are just different and do different things. Use what's appropriate for your needs and leave it at that.
I can say as a long time contributor to opensource software I am disgusted at reading the comments of blowhard 'enthusiasts' who denigrate the hard work and contributions of hundreds of people when they get in these pissing matches. I am friend with Linux kernel contributors and I can guarantee we don't flame each other in this manner.
At the moment I write this there are 297 comments mostly debating the merits of LLVM/Clang vs. GCC. There is not one mention of EGCS.
Fifteen years ago GCC was forked. A group of people we're frustrated with GCC and its leadership because they had contributions to make and talent to offer that was not welcome. They called their fork EGCS.
Why are we doing this? It's become increasingly clear in the course of hacking events that the FSF's needs for gcc2 are at odds with the objectives of many in the community who have done lots of hacking and improvment [sic] over the years.
The GCC you use today is EGCS. A few years later EGCS was adopted as GCC 2.95 after the merits of EGCS became undeniable.
Looks like we've come full circle. The cool kids are off in the weeds making cool stuff. Better stuff, and the `Powers That Be' are not interested. The `needs' of the FSF today are no longer in sync with the `needs' of the developers of today.
The bottom line is that GCC as it is with it's leadership, code base and license agenda doesn't cut it for those who have the talent, motivation and capital to create a tool chain that does cut it. You don't get to impede that, however righteous you think you are.
Freedom. Deal with it.
Lurking at the bottom of the gravity well, getting old
The problem with that argument is that what is best for the software and humanity is not clear cut. Most of the better software out there has significant corporate backing. Far too often, open source software falls into the trap of writing code that "works for me", where "me" is defined as the person who wrote it, yet tends not to "work for me", where "me" is defined as anyone else. Corporate backing tends to fix a lot of that because you have lots of "mes" working on the code, each of whom has a significant interest in making it work correctly and reliably (because they're getting paid to spend their time doing so). Any licensing requirements that are sufficiently onerous to scare away that corporate backing, therefore, tend to result in software of lesser quality.
IMO, the ideal situation is a BSD or similar license with the code owned by a non-profit organization. In this way, you have a reasonable assurance that the code won't suddenly get closed by its primary maintainer, and other companies are unlikely to want to close the code themselves because of the maintenance headaches of keeping a proprietary branch in sync with something that is regularly getting updated by others. However, companies are willing to work on the software and improve it because they don't have to worry about crossing some fuzzy line and getting sued.
Check out my sci-fi/humor trilogy at PatriotsBooks.
Have you ever actually tried to parse a declaration reliably using regular expressions? I have. They just kept getting more and more horrible the more complex the input became, and at some point, after fixing a bunch of bugs, I concluded that it was an insane way of doing things, threw out all the regular expressions, and started over with a tokenizing parser.
Trust me when I say that parsing declarations with even Perl-compatible regular expressions, much less BRE or ERE, is not something you want to attempt if you value your sanity in the slightest.
Check out my sci-fi/humor trilogy at PatriotsBooks.
One reason for that is because the GCC team won't accept Apple's patches for new versions of Objective-C. Apple want to move Objective-C forward, GCC has become a barrier to that, so they support CLANG/LLVM development. The version of GCC included is simply for legacy support and will be removed in due course once CLANG support for C++ is good enough.
Clang/LLVM also gives you nifty stuff for interfacing with the IDE, far better compilation errors/warnings, faster compile times, etc.
I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
We don't have to guess which model works best, at this point we have historical data. Your model failed with respect to X. MIT created and maintained an X that they released via. the MIT license. All the UNIX vendors then took this MIT code and intermixed it with their custom code creating value add X's that were specific to their platform, and closed source. The effect was that the X that existed in the public domain was worthless for end users, and the X's that were worthwhile were closed. X itself couldn't progress because it fragmented so all the interesting stuff existed in other layers. Years later when there was a desire for a workable open X, the XFree86 project had to start, essentially from scratch and this took years. We still haven't gotten all the features that existed in those proprietary Xs 2 decades ago.
That is the classic example of why BSD style licensing doesn't work. The primary maintainer is not unchanging.
Conversely the GPL has a long history of successful multi corporate contributions over time. The historical data simply refutes your theory of what should work.
Far too often, open source software falls into the trap of writing code that "works for me", where "me" is defined as the person who wrote it, yet tends not to "work for me", where "me" is defined as anyone else. Corporate backing tends to fix a lot of that because you have lots of "mes" working on the code, each of whom has a significant interest in making it work correctly and reliably (because they're getting paid to spend their time doing so).
I don't see this working in the real world, e.g. with Android. Corporate backing tends to push code of low quality (cf. the plethora of bugs that were fixed when the Android specific code was put in the upstream Linux kernel) because it was written quickly due to the corporation feeling the pressure from its competitors, and because its developers are paid for the time they spend coding; their interest is focused on solving the corporation's own problems (a corporation is a very big and selfish "me") with no regards to the effect that their solution will have on others' problems (cf. what happened with Apple and CUPS). And when a corporation has moved to the next product, they have no interest whatsoever for either the old code itself or its users (cf. what happens every time a new release of Android is revealed and users would like to upgrade, but they can't because of the binary blobs or forked code).
Any licensing requirements that are sufficiently onerous to scare away that corporate backing, therefore, tend to result in software of lesser quality.
This is not what I'm seeing with GPL projects such as Linux and the GCC. I think that the code quality of an open source project depends more on the community that it's able to gather than its license. But even if it we assume it's so, then the problem lies with the FUD about the license rather than in the license itself. FUD that I find in your comment, too:
they don't have to worry about crossing some fuzzy line and getting sued.
No company has ever been sued because of "crossing some fuzzy line". A couple of companies were sued because they absolutely refused to put a tarball on an FTP site despite the fact that the authors of that code had tried to convince them to do so for years. In comparison, Google is getting sued to hell because of BSD-licensed code. The truth is that no license will make you safe from copyright/patent trolls.