Re: GLIBC 2.0.5c changes...
On Friday, 14 Nov, Michael Alan Dorman wrote:
> It's already being done---we're getting the right kernel headers.
With libc6.1-dev? Well, there are some minor differences between Alpha
and other versions of include/linux, but it does not seem to be a
problem. I hope you are right. But I am not 100% sure. Especially
when it comes to kernel compilation.
> Symlinks are bad, and Linus himself has come out against them.
> They're no longer necessary for kernel compilation, and the only thing
> resembling a compatability guarantee we get from Linus is for upwards,
> not downwards compatability.
Right, and still there is no actual upwards compatibility.
> What if you're running a 2.1 kernel and I'm still running 2.0.31?
> libc should include the headers from the latest mainstream
> kernel---which is generally the latest stable kernel, although I
> wouldn't think it totally impossible for them to reflect a
> particularly common and stable late-model development kernel.
Good idea. That's true that we must be careful when compiling against
the latest kernel headers, but some packages (e.g., modutils) may
actually benefit from it (modutils would try the newer interface and
fall back on 2.0 kernels, which is very good, because this is the case
when the kernel interfaces are completely incompatible).
But this has nothing to do with my original point. I did not try to
advocate symlinks. I only was a bit worried that if we used
non-Alpha-patched headers to compile stuff on Alpha, we might
eventually end up with wrong results. As soon as it gets fixed in glibc
or kernel-headers, symlinks (on Alpha) will happily go away.
On Friday, 14 Nov, Michael Meskes wrote:
> kernel-headers is not obsolete at all. For instance there is no
> modversion.h in the glibc headers.
So, what is the story then? What goes into glibc and what goes into
kernel-headers? Why are kernel headers distributed between two
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
Trouble? e-mail to email@example.com .