Hi Will,

On 11/9/2016 1:16 PM, Will Deacon wrote:

On Wed, Nov 09, 2016 at 05:54:13PM +0000, Mark Rutland wrote:

quoted

On Fri, Oct 28, 2016 at 04:50:13PM -0400, Neil Leeder wrote:

quoted

+	struct perf_event *events[MAX_L2_CTRS];
+	struct l2cache_pmu *l2cache_pmu;
+	DECLARE_BITMAP(used_counters, MAX_L2_CTRS);
+	DECLARE_BITMAP(used_groups, L2_EVT_GROUP_MAX + 1);
+	int group_to_counter[L2_EVT_GROUP_MAX + 1];
+	int irq;
+	/* The CPU that is used for collecting events on this cluster */
+	int on_cpu;
+	/* All the CPUs associated with this cluster */
+	cpumask_t cluster_cpus;

I'm still uncertain about aggregating all cluster PMUs into a larger
PMU, given the cluster PMUs are logically independent (at least in terms
of the programming model).

However, from what I understand the x86 uncore PMU drivers aggregate
symmetric instances of uncore PMUs (and also aggregate across packages
to the same logical PMU).

Whatever we do, it would be nice for the uncore drivers to align on a
common behaviour (and I think we're currently going the oppposite route
with Cavium's uncore PMU). Will, thoughts?

I'm not a big fan of aggregating this stuff. Ultimately, the user in the
driving seat of perf is going to need some knowledge about the toplogy of
the system in order to perform sensible profiling using an uncore PMU.
If the kernel tries to present a single, unified PMU then we paint ourselves
into a corner when the hardware isn't as symmetric as we want it to be
(big/little on the CPU side is the extreme example of this). If we want
to be consistent, then exposing each uncore unit as a separate PMU is
the way to go. That doesn't mean we can't aggregate the components of a
distributed PMU (e.g. the CCN or the SMMU), but we don't want to aggregate
at the programming interface/IP block level.

We could consider exposing some topology information in sysfs if that's
seen as an issue with the non-aggregated case.

Will

So is there a use-case for individual uncore PMUs when they can't be 
used in task mode or per-cpu?

The main (only?) use will be in system mode, in which case surely it 
makes sense to provide a single aggregated count?

With individual PMUs exposed there will be potentially dozens of nodes 
for userspace to collect from which would make perf command-line usage 
unwieldy at best.

Neil
-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm 
Technologies Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project.