Shared publicly  - 
 
And people wonder why kernel maintainers are grumpy.

This patch is why, someone thinks they are "smarter" than the kernel and create an empty function to silence the kernel warnings that happen if they didn't provide a callback function. The programmer thinks they are smarter than the kernel and have silenced the nasty messages it was spitting out at them.

It's as if they think the kernel is doing this for no reason, and that an empty callback function really is the solution. Luckily, the kernel documentation says that I am allowed to publicly make fun of anyone who does this, so I don't feel bad about doing it.

But still, if you are a programmer, did you ever wonder about why the kernel/compiler/whatever is giving you a warning message? It is trying to help you out, some developer took their time to create that message and the logic to trigger it to be printed out. They didn't do it "just for fun", they were trying to save yourself a whole lot of problems in the end. By circumventing these types of messages, you don't solve the correct problem.

To quote the old IBM phrase, "THINK".
To which I'll add "or you will be mocked."
115
27
Daniel Bishop's profile photoJulian Calaby's profile photoAdrien Tougas's profile photoNishanth Menon's profile photo
20 comments
 
And I always thought you guys would just make the messages to annoy everyone. 😉 (Obviously I’m kidding, in case you can’t see the modernish Unicode Emoticon.)
 
Publicly making fun of people is half the fun of open source programming.

In fact, the real reason to eschew programming in closed environments is that you can't embarrass people in public.
 
A sense of humor is mandatory for open source programmers. And monkey suits.
 
Not only making fun, but also showing up there are seudo-programmers that mess up on real kernel hackers daily work.
Kudos for Greg!
 
What part pf the kernel tree should i mess up so i can have the gods of linux notice me and mention me in public? :-D
 
+Linus Torvalds I totally agree. Some software is only kept "closed source" so noone out there will publicly mock the developers, which would probably have a hard time finding a job afterwards.
This saves jobs, for sure, but hinders overall quality.
 
+Max Seifert i agree completely. I have seen commercial code that cost more than $200000 that would have made me cry if i had paid for it. In particular one developer who luckily always put a comment with his name wherever he 'fixed' code so at least it was easy to find the worst bits of code...

Sadly it wasn't open source so i don't think i should mention even the name of his company.
 
+birger monsen, the whole point of this is naming and shaming, if you don't mention the names, you're doing it wrong =)
 
+Max Seifert You can basically look at any router firmware that has been forced opensource by GPL. I don't think all those companies are violating the GPL because they have innovation to hide but because their programmers are embarrassed by their own code.
 
+Julian Calaby i would have to be fair. Since i cannot refer to the offending code i cannot give names either. It only works for open source code as +Linus Torvalds says above
 
The neckbeard is strong with this one.
 
Any people wonder why Windows and OSX are getting all of the good developers. Last time someone mocked me about my code, I deleted their server hosting their code.
 
I realize this is a public G+ post, but, +Linus Torvalds , do you mind if I include your comment from the top of the thread in my open-source projects in the future?
 
+birger monsen Fair enough, I'd love to mock a particular company's poorly written web services, but I work for them which makes it awkward.
 
+John Nelson In the same paragraph you gloat about being childish and scold someone else for their behavior. Beautiful!
 
Its called hypocrisy. Welcome to 'MURIKA
 
So.. we need the ksyms function to get the size of a function, and extend the warning to empty ones ... not all that hard to do.
 
+Arjan van de Ven I had a hack in -mm for a long time that did something like that, but it only worked on x86 builds for some odd reason, so I could never push it in. I'll gladly take a patch that adds this if it is possible.
Add a comment...