Re: [PATCH] move put_task_struct() reaping into a thread [Re: 2.6.18-rt1]

[hui]

On Wed, Sep 27, 2006 at 10:57:12AM +0200, Ingo Molnar wrote:
> * Bill Huey <billh@xxxxxxxxxxxxxxxxx> wrote:
> 
> > Because the conversion of memory allocation routines like kmalloc and 
> > kfree aren't safely callable within a preempt_disable critical section 
> > since they were incompletely converted in the -rt. [...]
> 
> they were not 'incompletely converted' - they are /intentionally/ fully 
> preemptible.

What I meant by "incompletely converted" is that the allocators could be
made more safe in non-preemptible scenarios under -rt. It's potentially
a valuable thing to have since GFP_ATOMIC semantics already exist in the
current allocators and a newer category could be added as a new feature of
that allocator for those scenarios. I'm happy dequeuing things off of my
own free list, but that's just me.

-rt semanatics created a couple of new locking scenarios that the previous
kernel didn't really have to address. That's all that I meant by that. :)

bill

-
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/