> On Jul 11, 2015, at 10:44 AM, Frederic Weisbecker wrote:
>
>> On Fri, Jul 10, 2015 at 03:37:25PM -0400, Chris Metcalf wrote:
>> Normally the tilegx networking shim sends irqs to all the cores
>> to distribute the load of processing incoming-packet interrupts,
>> so that you can get to multiple
On Fri, Jul 10, 2015 at 03:37:25PM -0400, Chris Metcalf wrote:
> Normally the tilegx networking shim sends irqs to all the cores
> to distribute the load of processing incoming-packet interrupts,
> so that you can get to multiple Gb's of traffic inbound.
>
> However, in nohz_full mode we don't wan
On Fri, Jul 10, 2015 at 07:06:23PM -0400, Chris Metcalf wrote:
> On 7/10/2015 6:45 PM, Josh Cartwright wrote:
> >>+static inline const struct cpumask *housekeeping_cpumask(void)
> >>>+{
> >>>+#ifdef CONFIG_NO_HZ_FULL
> >>>+ if (tick_nohz_full_enabled())
> >>>+ return housekeeping_mask;
>
On 7/10/2015 6:45 PM, Josh Cartwright wrote:
+static inline const struct cpumask *housekeeping_cpumask(void)
>+{
>+#ifdef CONFIG_NO_HZ_FULL
>+ if (tick_nohz_full_enabled())
>+ return housekeeping_mask;
>+#endif
Just a small comment:
We can take these checks out from under a #ifdef C
On Fri, Jul 10, 2015 at 03:37:25PM -0400, Chris Metcalf wrote:
> Normally the tilegx networking shim sends irqs to all the cores
> to distribute the load of processing incoming-packet interrupts,
> so that you can get to multiple Gb's of traffic inbound.
>
> However, in nohz_full mode we don't wan
Normally the tilegx networking shim sends irqs to all the cores
to distribute the load of processing incoming-packet interrupts,
so that you can get to multiple Gb's of traffic inbound.
However, in nohz_full mode we don't want to interrupt the
nohz_full cores by default, so we limit the set of cor
