You stole the words from my mouth. Goddamn they are taking long to come with the final product. :)

Interestingly, the Finnish stub of Nokia that was left, is doing fine. They still have a feasible telecommunication networks business.

Yes, but pinpointing the problem in GCC source code is another story.

Yeah, that can be a slight problem... :)

Try fixing a bug in an OSS project sometimes, if you haven't. It can take ages to familiarize yourself with the project to be able to pinpoint the problem.

You can't just go and fix every program, even if they are open source. It takes a lot of time to familiarize one with the code base. That is probably limiting Linus here too. His bug report is very precise, and that will greatly ease the job of the GCC developer who actually fixes the bug.

Actually Visual Studio has Makefile support. :)

The problem with Linux and OSS in general is that these kind of weird problems can pop up any time and are partly the responsibility of the general public to test and report. With Windows and Mac, the extensive quality assurance catches a lot of this stuff and I do not have to worry about it in the first place. I am not saying that even commercial OSs are perfect, but they are much more smooth sailing. I still share your argument that with closed software, it is almost impossible to reach the engineers if there is a problem.

Nice article, that should explain a bunch of things to me.

Troll, troll, yeah, I'm so troll...

Another worrying thing which Linus says there is:

"The problem is that these things are a bitch to debug - they turn into these completely impossible kernel oopses or corruption, and we were just very lucky that this one case happened to be repeatable and pinpoint for two people. Are there others? We have no way of knowing.."

These days I interpret Linus' "meltdowns" just as some funny nerd rage. He uses that technique to strongly underline the importance of his point, it's never real anger. Often there's a dash of humor in the mix, such as in this case the comment "that compiler shouldn't have been allowed to graduate from kindergarten".

Can you explain what is the %rbp, and why are its offsets negative? If I set values 1, 2, 3, 4 to a bunch of variables, I get this assembly code:

movl $1, -4(%rbp)
movl $2, -8(%rbp)
movl $3, -12(%rbp)
movl $4, -16(%rbp)

A Beowulf cluster generally consists of "ghetto hardware", for example cheap custom-built PCs or Raspberry Pis. If these are professional servers, they do not classify as such.