Re: [PATCHv12 0/3] rdmacg: IB/core: rdma controller support

[Parav Pandit <pandit.parav-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>]

Hi Liran,

On Fri, Nov 4, 2016 at 10:36 AM, Liran Liss <liranl-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> wrote:
>> From: Parav Pandit [mailto:pandit.parav-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org]
>
>>
>> On Fri, Nov 4, 2016 at 10:22 AM, Liran Liss <liranl-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> wrote:
>> >> From: Parav Pandit [mailto:pandit.parav-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org]
>> >
>> >> >
>> >> > A global HCA metric is indeed in the right direction.
>> >> > However, rethinking this, I think that we should specify the metric
>> >> > in terms of
>> >> RDMA objects rather than percentage.
>> >> > Basically, any resource that consumes an IDR is charged.
>> >> >
>> >> If metric definition is based on RDMA objects (count) and not based
>> >> on percentage, how would user specify the metric without really
>> >> specifying object type.
>> >> Current patch defines the metric as absolute numbers and objects as well.
>> >>
>> >
>> > That is the requested change. The absolute number would account for any
>> object allocation. We won't distinguish between types.
>> > Only a single counter (per device).
>> >
>>
>> In that case ucontext deserve a additional count. Because that is handful in
>> range of 256 to 1K.
>> If we give absolute consolidated number as 2000, one container will allocate all
>> the doorbell uctx and no other container can run.
>> Percentage works for this particular case.
>>
>
> Hmm..
> I guess that you are right.
>
> So we can add another count for "HCA handles",
I prefer this. This keeps it vendor agnostic and clean if we don't go
percentage route.
Would indirection table also fall in this category?

> or alternatively, each provider will restrict the number of handles per device to a reasonable small number (which
> won't be treated as one of the "HCA resources").
This would require vendor drivers to get the understanding of cgroup
object and pid and that breaks the modular approach. I like to avoid
this.

> Typically, a process shouldn't need to open more than a single handle...
Right. well behaved application won't do multiple handles.


>
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html