Oops sorry I take that back, gcc3 just produces different output...
Never mind then. Must get coffee.
Jacob
On 10/5/05, Jacob Gorm Hansen <jacobg@xxxxxxx> wrote:
> On 10/5/05, Petersson, Mats <mats.petersson@xxxxxxx> wrote:
>
> > I created a function:
> >
> > int foo(char *s)
> > {
> > if (strlen(s)) return 1; else return 0;
> > }
> >
> > Compiled with gcc 4.0.0 as:
> > gcc -O3 -s x.c (or gcc -O2 ...)
> > Gives the following code:
> >
> > foo:
> > xorl %eax, %eax
> > cmpb $(0), (%rdi)
> > sete %al
> > ret
>
> Cool, I am surprised gcc4 is able to fix this. gcc 3.3.4 which I am
> using is not, even with -O3. How does gcc4 behave without -O3?
>
> foo:
> push %ebp
> mov %esp,%ebp
> mov 0x8(%ebp),%edx
> pop %ebp
> cmpb 0ドルx0,(%edx)
> setne %dl
> movzbl %dl,%eax
> ret
>
> I guess I only reacted to this because I have seen worse examples
> resulting from the "strlen is constant-time" assumption (people
> iterating over a string in a while(strlen(s)) loop), and because I was
> bored with trying to get block devices running with xenbus. :-)
>
> Jacob
>
--
Save time and bandwidth with EDelta: http://www.diku.dk/~jacobg/edelta/
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel