[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200722072943.GD9290@linux.vnet.ibm.com>
Date: Wed, 22 Jul 2020 12:59:43 +0530
From: Srikar Dronamraju <srikar@...ux.vnet.ibm.com>
To: Gautham R Shenoy <ego@...ux.vnet.ibm.com>
Cc: Michael Ellerman <michaele@....ibm.com>,
linuxppc-dev <linuxppc-dev@...ts.ozlabs.org>,
LKML <linux-kernel@...r.kernel.org>,
Ingo Molnar <mingo@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
Valentin Schneider <valentin.schneider@....com>,
Nick Piggin <npiggin@....ibm.com>,
Oliver OHalloran <oliveroh@....ibm.com>,
Nathan Lynch <nathanl@...ux.ibm.com>,
Michael Neuling <mikey@...ux.ibm.com>,
Anton Blanchard <anton@....ibm.com>,
Vaidyanathan Srinivasan <svaidy@...ux.ibm.com>,
Jordan Niethe <jniethe5@...il.com>
Subject: Re: [PATCH v2 09/10] Powerpc/smp: Create coregroup domain
> > @@ -1386,6 +1421,9 @@ int setup_profiling_timer(unsigned int multiplier)
> >
> > static void fixup_topology(void)
> > {
> > + if (!has_coregroup_support())
> > + powerpc_topology[mc_idx].mask = cpu_bigcore_mask;
> > +
>
> Shouldn't we move this condition after doing the fixup for shared
> caches ? Because if we have shared_caches, but not core_group, then we
> want the coregroup domain to degenerate correctly.
>
Currently we aren't consolidating, and hence the order doesn't matter for
degeneration. However even if in future, if we want to consolidate the
domains before calling set_sched_topology(), this order would be ideal.
>
> > if (shared_caches) {
> > pr_info("Using shared cache scheduler topology\n");
> > powerpc_topology[bigcore_idx].mask = shared_cache_mask;
>
--
Thanks and Regards
Srikar Dronamraju
Powered by blists - more mailing lists