[discuss] Re: [PATCH 5/39] NLKD/x86-64 - early/late CPUup/down notification

[Jan Beulich]

>>> Andi Kleen <ak@xxxxxxx> 10.11.05 14:10:16 >>>
>On Wednesday 09 November 2005 15:01, Jan Beulich wrote:
>> x86_64-specific part of the new mechanism to allow debuggers to
learn
>> about starting/dying CPUs as early/late as possible.
>
>Please just use the normal notifier chains instead (CPU_UP, CPU_DOWN,

>register_cpu_notifier). I don't see much sense to have two different 
>mechanisms to do the same thing. While they might be not as
early/late
>as your mechanism I think the users of your debugger can tolerate
that.

Assuming you mean CPU_ONLINE and CPU_DEAD. But no, I don't really like
this. The most significant difference is that the existing notifications
are not sent on the starting CPU, but on the one it got started from.
The point in time is only the second reason for not using these.

Jan
-
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/