Re: [PATCH v2] irqchip/gic-v3-its: Balance initial LPI affinity across CPUs

[Marc Zyngier]

Hi John,

On 2020-03-12 15:41, John Garry wrote:
> On 12/03/2020 11:55, Marc Zyngier wrote:
>
> Hi Marc,
>
> > When mapping a LPI, the ITS driver picks the first possible
> > affinity, which is in most cases CPU0, assuming that if
> > that's not suitable, someone will come and set the affinity
> > to something more interesting.
> >
> > It apparently isn't the case, and people complain of poor
> > performance when many interrupts are glued to the same CPU.
> > So let's place the interrupts by finding the "least loaded"
> > CPU (that is, the one that has the fewer LPIs mapped to it).
> > So called 'managed' interrupts are an interesting case where
> > the affinity is actually dictated by the kernel itself, and
> > we should honor this.
> >
> > Reported-by: John Garry <john.garry@xxxxxxxxxx>
> > Link: 
> > https://lore.kernel.org/r/1575642904-58295-1-git-send-email-john.garry@xxxxxxxxxx
> > Signed-off-by: Marc Zyngier <maz@xxxxxxxxxx>
> > Cc: John Garry <john.garry@xxxxxxxxxx>
> > Cc: Ming Lei <ming.lei@xxxxxxxxxx>
> > ---
> > Reviving this at John's request.
>
> Thanks very much. I may request a colleague test this due to possible
> precautionary office closure.

Huh. Not great... :-(

>  The major change is that the
> > affinity follows the x86 model, as described by Thomas.
>
> There seems to be a subtle difference between this implementation and
> what Thomas described for managed interrupts handling on x86. That
> being, managed interrupt loading is counted separately to total
> interrupts per CPU for x86. That seems quite important so that we
> spread managed interrupts evenly.

Hmmm. Yes. That'd require a separate per-CPU counter. Nothing too 
invasive
though. I'll roll that in soon. I still wonder about interaction of 
collocated
managed and non-managed interrupts, but we can cross that bridge later.

> > I expect this to have an impact on platforms like D05, where
> > the SAS driver cannot use managed affinity just yet.
>
> I need some blk-mq and SCSI changes to go in first to improve the
> interrupt handling there, hopefully we can make progress on that soon.

That'd be good.

Thanks,

        M.
--
Jazz is not dead. It just smells funny...