On Tue, 11 Sep 2007, Pierre Habouzit wrote:
quoted text
>
>   It has nothing to do with latest.

But it does.

The "latest" being the compiler.

A lot of people have old compilers, possibly even ones where there *are* 
no compiler versions that support C99.

quoted text
> I feel that if for some reason transport_ops need a new function

The best way to handle that is to make sure you use ANSI C (which we *do* 
depend on), and strict prototypes everywhere. Then, any breakage will not 
be silent, because different functions will have different prototypes.

Yes, C99 structure initializers are a good thing, and we use them in the 
linux kernel, but the kernel can make more assumptions about the compiler 
than a random user land tool can. So git should be more conservative than 
that.

It wasn't that long ago that people avoided even ANSI C, on the grounds 
that some (totally broken crap) OS's shipped with just a K&R compiler by 
default, and you had to pay for the "fancy" ANSI C compiler. Now, *that* 
kind of ass-backwards compatibility is just stupid in this day and age 
(especially since the advantages of ANSI over K&R are just so huge), but 
at assuming everybody has C99 is not really realistic yet, and the things 
C99 brings aren't quite worth breaking old compiler setups for.

(But we may be getting close to being able to assume C99. It's getting 
pretty common).

			Linus
-
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html