Re: [announce] new tree: "fix all build warnings, on all configs"

[Linus Torvalds]

On Sat, 18 Oct 2008, Ingo Molnar wrote:
> 
> thx, you are right - fixed it via the patch below.

Hell no.

The old code was correct. Your code is shit. And you didn't fix 
_anything_.

>  	case 5:
>  		*(int *)to = *(int *)from;
> -		*((short *)to + 3) = *((short *)from + 3);
> +		*((char *)(to + 3)) = *((char *)(from + 3));
>  		return to;

Are you just making changes by randomly inserting and deleting characters 
until you don't see warnings? Or what?

That thing is supposed to be a 5-byte memcpy. Not a "take a random byte 
from a random location and move it to another random location". That would 
be "randcpy()", not "memcpy()".

I don't want to see obvious and shitty crap like this. I don't want to 
pull from people who write code with some "random walk" algorithm.

F*ck me, what's wrong with you people? THAT CODE WAS NOT BUGGY. If it 
causes a warning, it is because SOME CALLER used a 5-byte memcpy() on 
something that gcc thought was just four bytes in size.

Ingo, I'm not going to pull _anything_ from you.

			Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/