Re: [RFC, Announce] Unified x86 architecture, arch/x86

[Glauber de Oliveira Costa]

On 7/21/07, Andi Kleen <ak@xxxxxxx> wrote:
> [of which several just #include <asm-generic/foo.h] ?
>
> I suppose msidef.h and hypertransport.h should be shared agreed;
> for the others it is not obvious. spinlock_types will likely fork soon;
> it used to be different in the past already and will be again.
Why will it fork? I don't think it will ever happen that the trees
will have large pieces that _has_ to be different one from the other.
So if it's forking to achieve some benefits, why can't i386 get the
benefits too? I think this is the whole point here.

Surely as it is today (and just because it wasn't merged earlier and
the past!), the x86_64 tree has a bunch of things that are quite
better structured than the i386 (and maybe vice-versa, but I must
admit that unlike Steven Roasted, I like the x86_64 a lot more). But
in the long term,  it tends to just get the best of each picked up.

And oh yeah, i386 is older, has a lot more corner cases, but even if
it does count against the merge, we have a net win at the end.


--
Glauber de Oliveira Costa.
"Free as in Freedom"
http://glommer.net

"The less confident you are, the more serious you have to act."
-
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/