FreeBSD 9.1: LLVM/Clang Battling GCC

[0xBADCODE]

And I still don't understand the value of benchmarks like that. The same with graphics driver tests. Pages and pages of graphs showing performance under poor conditions instead of showing what the drivers (or compilers) are actually capable off.

Most users would never adjust defaults. "Defaults will prevail". So that's what >90% of users will actually be able to get from system. Hence it's completely fair and good idea to compare default systems without tuning.

[systemd rulez]

And it's better too see bad graphs immediately rather than install OS, waste a lot of time and then get surprised by shitty performance.

Classic BSD

[Vim_User]

Quite simple: I'm unwilling to make any discounts and if someone loses in OpenMP here and now by a factor of 3, let's admit this problem and fix it if you can. Just hiding problematic tests is lame. And as for me I'm using OSes today. I don't want to wait for June or whatever to get adequate performance. And it's better too see bad graphs immediately rather than install OS, waste a lot of time and then get surprised by shitty performance.

I agree that lack of support for something like OpenMP should not be hidden. But what is the point of benchmarking a not supported feature in a comparison? Just a waste of time, one could simply state that OpenMP is not supported instead.

[demonkoryu]

Christ, are you guys bashing clang/BSD serious or am I just dropping into an exchange of in-jokes? I mean, this article is only relevant to you if you a) are running BSD and b) are interested in the performance impact of the clang switch. If you are not, then why are you wasting your and everyone else's time here debating non-issues?

[archibald]

Christ, are you guys bashing clang/BSD serious or am I just dropping into an exchange of in-jokes? I mean, this article is only relevant to you if you a) are running BSD and b) are interested in the performance impact of the clang switch. If you are not, then why are you wasting your and everyone else's time here debating non-issues?

They're just trolling. I've not idea why they do it.

[OpenSLOWlaris]

Christ, are you guys bashing clang/BSD serious or am I just dropping into an exchange of in-jokes? I mean, this article is only relevant to you if you a) are running BSD and b) are interested in the performance impact of the clang switch. If you are not, then why are you wasting your and everyone else's time here debating non-issues?

This benchmarks with a new version of LLVM vs an old outdated version of GCC is often misused by BSD supporters to claim that GCC is crap and thus the GPL is also crap. BSD supporters do not use this article to see that impact of the switch to clang as that would be thought crime to them. Also, anyone concerned with performance (or security or portability for that matter) would never use any BSD crap. So this article will never be used by BSD supports for what you claim. No they use it for trolling.

The people you accuse of bashing clang/BSD are good willed people trying to stop that from happening. BSD supporters are vile and blatent.

They're just trolling. I've not idea why they do it.

The real trolls in this thread are brainless BSD supporters like you, BitLight, vermanden, phkbsd, VertexSymphony and Vim_User. If there is one thing that can be improve in the conversation, it be your posts be invisible to others in threads similiar to this one.

[Vim_User]

thus the GPL is also crap.

Nobody but you is talking about the GPL here.

Besides that, after your pathetic and criminal outburst (accusing the OpenBSD people to be the Boston bombers) I am surprised that you are still allowed to post on this forum.

[Sergio]

This benchmarks with a new version of LLVM vs an old outdated version of GCC is often misused by BSD supporters to claim that GCC is crap and thus the GPL is also crap. BSD supporters do not use this article to see that impact of the switch to clang as that would be thought crime to them. Also, anyone concerned with performance (or security or portability for that matter) would never use any BSD crap. So this article will never be used by BSD supports for what you claim. No they use it for trolling.

The people you accuse of bashing clang/BSD are good willed people trying to stop that from happening. BSD supporters are vile and blatent.



The real trolls in this thread are brainless BSD supporters like you, BitLight, vermanden, phkbsd, VertexSymphony and Vim_User. If there is one thing that can be improve in the conversation, it be your posts be invisible to others in threads similiar to this one.

Yes sir, you are back... Keep this thread alive so more people can be aware of the great things happening with FreeBSD. Thanks!

[gamerk2]

The irony here is that, compared to what we have in the Windows world, no one bothers with GCC. Slightly slower then MSVC, and a hack of a lot harder to, you know, CODE with.

http://www.behardware.com/articles/8...itectures.html



As a developer, if I had to code for a non-Windows platform, I'd use LLVM/CLANG in a heartbeat over GCC, especially once it gains OpenMP support later this year.

[ArneBab]

As a developer, if I had to code for a non-Windows platform, I'd use LLVM/CLANG in a heartbeat over GCC, especially once it gains OpenMP support later this year.

See, that’s why I do not like the focus the OpenMP tests put on OpenMP: They create a hype on OpenMP which establishes the feeling that LLVM would be so much better than GCC if it just implemented OpenMP - which is simply a feature which will come at some point.

That’s like establishing a binary decision point when people should switch instead of showing the different aspects of choosing a compiler.