Re: Hotplug CPU printk issue

[Rusty Russell <rusty@rustcorp.com.au>]

On Thu, 2005-05-26 at 12:32 -0700, David Mosberger wrote:
> >>>>> On Wed, 25 May 2005 22:52:04 -0700, Andrew Morton <akpm@osdl.org> said:
> 
>   Andrew> Shaohua Li <shaohua.li@intel.com> wrote:
> 
>   >> > Please confirm that we in fact do not want to allow downed CPUs to
>   >> > print things, then send a patch.
>   >> Yep. In the cpu hotplug case, per-cpu data possibly isn't initialized
>   >> even the system state is 'running'. As the comments say in the original
>   >> code, some console drivers assume per-cpu resources have been allocated.
>   >> radeon fb is one such driver, which uses kmalloc. After a CPU is down,
>   >> the per-cpu data of slab is freed, so the system crashed when printing
>   >> some info.
> 
>   Andrew> hm, that certainly sounds sane, but I do recall there were
>   Andrew> reasonable-sounding reasons why the ia64 guys wanted
>   Andrew> printk-on-a-down-CPU to work.  Hopefully David can remember
>   Andrew> what the problem was so we can find a more thorough fix.
> 
> I don't recall having submitted such a patch.  According to the bk
> log, it was Rusty who added the !system_running check (which was later
> changed to system_state != SYSTEM_RUNNING).
> 
> The changelog only says:
> 
>  "- Allow printk on down cpus once system is running"

IIRC it was a great aid to debugging hotplug CPU, and there seemed no
reason to ban it.  I mean, you have to be running code on a down CPU,
which implies you're in arch code or you've done something wrong... I
don't have religious attachment to it, however!

Cheers,
Rusty.
-- 
A bad analogy is like a leaky screwdriver -- Richard Braakman

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/