Re: New filesystem for Linux

[Oleg Verych <olecom@flower.upol.cz>]

On Fri, Nov 03, 2006 at 08:08:24PM +0100, Adrian Bunk wrote:
> On Fri, Nov 03, 2006 at 06:36:09PM +0100, Oleg Verych wrote:
> > On Fri, Nov 03, 2006 at 06:09:39PM +0100, Mikulas Patocka wrote:
> > > >In gmane.linux.kernel, you wrote:
> > > >[]
> > > >>From: Andrew Morton <akpm@osdl.org>
> > > >>
[^0] > > > >>As Mikulas points out, (1 << anything) won't be evaluating to zero.
> > > >
> > > >How about integer overflow ?
> > > 
> > > C standard defines that shifts by more bits than size of a type are 
> > > undefined (in fact 1<<32 produces 1 on i386, because processor uses only 5 
> > > bits of a count).
> > ,--
> > |#include <stdio.h>
> > |int main(void) {
> > |	unsigned int b = 1;
> > |
> > |	printf("%u\n", (1 << 33));
> > |	printf("%u\n", (b << 33));
> > |	return 0;
> > |}
> > |$ gcc bit.c && ./a.out
> > `--
> > 
> > There *is* difference, isn't it?
> 
> It's undefined, and the results with your example depend on the gcc 
> version and optimization level.
> 
> E.g. with gcc 4.1, there is *no* difference any more if you turn on 
> optimization.

Sure it is. And it is *zero*, not is stated in [^0].
,--
|olecom@flower:/tmp$ gcc --version
|gcc (GCC) 4.1.2 20060901 (prerelease) (Debian 4.1.1-13)
`--

Hmm. Did i spend more on uC C than PC C? Seem like yes.
So, pay no nevermind, please.

> cu
> Adrian
> 
____