Linus Torvalds On GCC 4.9: Pure & Utter Crap

[log0]

It is not sympathy he is lacking, but wisdom. The knowledge, which comes with experience. But it is like you say. He should know from his own experience how much work it is to run a project of this size, the problems it creates, and what it feels like when somebody comes a long and dumps an obviously stupid opinion on it. It is not the first time though and his rages are getting into the news. People will start to wonder if he is really needed to keep Linux alive or if it will continue without him. And frankly, I do not see why the Linux kernel project would not continue without him.

I am only worried about the big companies that are using Linux more and more often. It is no fun explaining to your bosses how Linux makes everything better, when its creator is developing such ludicrous behaviour and is making news with it.

Lol, as far as I can remember Linus has always been know for being very "direct", no idea why you are having a problem with his behavior now.

Can't handle it? Feel free to stick with microsoft products then. I've heard the recent layoff mail from microsoft has been an epitome of politically correct corporatre speak, blergh...

[milkylainen]

Ptss..

First off.
This is just ridiculous. Posting a GCC bug and Torvalds whining about GCC? I think this gets picked as an article because of Michaels utterly biased position against GCC. Clang/LLVM/whatever is oooh so shiny! GCC broken, bah, bleh etc.
This mantra repeats itself in every article posted by Michael.

Also:
Linus rants about a million things all the time. Right or wrong, this is his style. GCC getting broken the last couple of releases is nothing new.
They have been feeling the "OOOH SHINY" pressure from Clang/LLVM for quite some time now. This leads to more flashy features getting pushed with code generation correctness (more bugs introduced) taking a back seat.

In the end, this leads to some idiot bugs Linus is complaining about. In this case I couldn't agree more. I have also been pushing the same mantra about compiler correctness. It takes precedence over all shiny features. I could not care less about speedy compilers or blaha features if it does not get code generation right (adherence to flags, abi's etc).

Debugging your broken output is a nightmare coming true. It creates an insecurity about the generated code. If it broke this, what else did it break I have yet to discover?

[sdack]

Lol, as far as I can remember Linus has always been know for being very "direct", no idea why you are having a problem with his behavior now.

Can't handle it? Feel free to stick with microsoft products then. I've heard the recent layoff mail from microsoft has been an epitome of politically correct corporatre speak, blergh...

One wants to be taken serious, but nobody will when one is being rude, stupid and insulting. Mutual respect is the foundation of every good relationship.

Imagine you would be working on a project (and perhaps you are) and some guy sends you a bug report full of insults, but otherwise lacks insights into how your project handles bugs. Would you care for his report? Or would you rather tell the guy to go f' himself? I am sure it is the later, because you also take pride in your work. Or would a lot of insults make you roll over?

I am going to stay respectful here and I hope to get an honest answer, but if it helps for you to understand the problem can I give you as many insults as you need (or until the moderation stops me). *lol*

[dungeon]

mpv gets new release today... Instead users to read something about that, we have once again Linus said something about *shit*, it say *shit* and *crap* every *shity* day .

And i am shiting every day, that is healty .

[blackiwid]

the problem here is, he is not some random guy nobody knows, even then u dont always get good reactions and some people call u troll if u try to troll, like he did.

He should maybe learn taht he is not some random guy, and that he at least tries to troll less offensive. your project sucks, is also trolling but less offensive or more constructive than your project is utter shit or something like that.

I mean u can say such stuff to nvidia or other real asholes that make bad stuff out of pure evilness, but not against projects with people trying doing their best, and when their results are not sooo good it is because all the trolls that know how to make it better dont send patches or something.

But it seems his fame makes him even rant more, it seems a bit that its gone to his head a bit.

[carewolf]

From my point of view gcc is getting worse with every single major version (meaning the "x" in 4.x). Since 4.6 (which was quite decent IIRC), I have been facing lots of miscompilations caused by gcc emiting wrong code. With gcc 4.9 it even stopped accepting a valid C++ code from one of my projects. On the bright side ? after pointing out the commit which broke the compilation it was fixed quite promptly.

It only looks that way if you have a short memory. The early 3.x days and the unofficial 2.9x abominations was the golden age of miscompilations and internal compiler errors.

[jrch2k8]

the problem here is, he is not some random guy nobody knows, even then u dont always get good reactions and some people call u troll if u try to troll, like he did.

He should maybe learn taht he is not some random guy, and that he at least tries to troll less offensive. your project sucks, is also trolling but less offensive or more constructive than your project is utter shit or something like that.

I mean u can say such stuff to nvidia or other real asholes that make bad stuff out of pure evilness, but not against projects with people trying doing their best, and when their results are not sooo good it is because all the trolls that know how to make it better dont send patches or something.

But it seems his fame makes him even rant more, it seems a bit that its gone to his head a bit.

well, it is very relative, i mean yes when linus rants he is a lot like stan lee but is also true that when linus rants there is a very good reason too and normally after the explosion people tend to realize they were doing something wrong and introspect about it. In this case GCC guys may have let aside certain QA steps to push faster certain feature to make them look good against the shinny clang (dunno why tho, even today clang is too far away to even believe they can replace GCC in Linux) and linus won't allow that tendency to fuck up the kernel work.

On another subject i would love hear linus make a huge rant on the Glibc guys, like openssl i believe the monstrosity that has become today's glibc should be shaked to the foundations even if it means renumbering Glibc and break the current ABI.

[M@yeulC]

Hey

This

He wrote 1 email with a few sentences talking about how the code generation here sucks. If you look at the rest of the list and the bug reports, he also wrote about a dozen more where he was quite calm and helpful in trying to fix the problem.

This is another case of the media latching on to a sentence or two, taking them out of context, and then convincing people there is something noteworthy about it. If you've never said something sucks before in an email - well, i don't believe you.

And this :

First off.
This is just ridiculous. Posting a GCC bug and Torvalds whining about GCC? I think this gets picked as an article because of Michaels utterly biased position against GCC. Clang/LLVM/whatever is oooh so shiny! GCC broken, bah, bleh etc.
This mantra repeats itself in every article posted by Michael.

Also:
Linus rants about a million things all the time. Right or wrong, this is his style. GCC getting broken the last couple of releases is nothing new.
They have been feeling the "OOOH SHINY" pressure from Clang/LLVM for quite some time now. This leads to more flashy features getting pushed with code generation correctness (more bugs introduced) taking a back seat.

In the end, this leads to some idiot bugs Linus is complaining about. In this case I couldn't agree more. I have also been pushing the same mantra about compiler correctness. It takes precedence over all shiny features. I could not care less about speedy compilers or blaha features if it does not get code generation right (adherence to flags, abi's etc).

Debugging your broken output is a nightmare coming true. It creates an insecurity about the generated code. If it broke this, what else did it break I have yet to discover?

Calm down, guys. GCC is a huge project, certainly not a shitty one. Linus didn't insult the project, if I understand the sentence correctly (didn't read the mail), he said that some bug he diagnosed for an other person is caused by the compiler. The compiler did something it shouldn't have done, so the one used, the version used was shitty. He could have said the same about, say, someone who used a -RC kernel and reported a bug : "This/Your kernel is is pure and utter *shit*" with an advice : Upgrade/Downgrade/Wait next RC. This wouldn't have surprised anyone.
So, Michael, please stop making big titles about some sentences out of their context. Journalists love to do that, but I don't see any point in doing this but to spread more incomprehension about Linus, the Linux kernel and GCC.
Also, The picture is nice, but I don't think it's relevant. Less relevant than a shit picture, or a shitty one, I guess

(Also, I had to reactivate adblock because of fullscreen ads. Sorry for that, Michael. I appreciate your work.)

[nslay]

First off.
This is just ridiculous. Posting a GCC bug and Torvalds whining about GCC? I think this gets picked as an article because of Michaels utterly biased position against GCC. Clang/LLVM/whatever is oooh so shiny! GCC broken, bah, bleh etc.
This mantra repeats itself in every article posted by Michael.

Also:
Linus rants about a million things all the time. Right or wrong, this is his style. GCC getting broken the last couple of releases is nothing new.
They have been feeling the "OOOH SHINY" pressure from Clang/LLVM for quite some time now. This leads to more flashy features getting pushed with code generation correctness (more bugs introduced) taking a back seat.

In the end, this leads to some idiot bugs Linus is complaining about. In this case I couldn't agree more. I have also been pushing the same mantra about compiler correctness. It takes precedence over all shiny features. I could not care less about speedy compilers or blaha features if it does not get code generation right (adherence to flags, abi's etc).

Debugging your broken output is a nightmare coming true. It creates an insecurity about the generated code. If it broke this, what else did it break I have yet to discover?

I see how it is: If GCC is sucking, Michael is biased, when Clang is sucking, Clang sucks.

GCC has a bug and Linus called it crap *sniff*, Michael is totally biased against GCC *sniff*

Meanwhile, Michael continues to show benchmarks of Clang sucking with OpenMP (among other benchmarks). In those benchmarks, the same old people come out and cheer about how bad Clang is.

[warpedgeoid]

Torvalds and Gates act and behave nothing alike.



Linux and OpenBSD have been proving you wrong for over 20 years now.

Torvalds acts more like Jobs than Gates. That isn't necessarily a bad thing.