Re: [RFC v2] perf: Rewrite core context handling

[Ravi Bangoria <ravi.bangoria@amd.com>]

On 13-Jun-22 8:13 PM, Peter Zijlstra wrote:
> On Mon, Jun 13, 2022 at 04:35:11PM +0200, Peter Zijlstra wrote:
> 
>> @@ -3652,17 +3697,28 @@ static noinline int visit_groups_merge(s
>>  			.size = ARRAY_SIZE(itrs),
>>  		};
>>  		/* Events not within a CPU context may be on any CPU. */
>> -		__heap_add(&event_heap, perf_event_groups_first(groups, -1, NULL));
>> +		__heap_add(&event_heap, perf_event_groups_first(groups, -1, pmu, NULL));
>>  	}
>>  	evt = event_heap.data;
>>  
>> -	__heap_add(&event_heap, perf_event_groups_first(groups, cpu, NULL));
>> +	__heap_add(&event_heap, perf_event_groups_first(groups, cpu, pmu, NULL));
>>  
>>  #ifdef CONFIG_CGROUP_PERF
>>  	for (; css; css = css->parent)
>> -		__heap_add(&event_heap, perf_event_groups_first(groups, cpu, css->cgroup));
>> +		__heap_add(&event_heap, perf_event_groups_first(groups, cpu, pmu, css->cgroup));
>>  #endif
>>  
>> +	if (event_heap.nr) {
>> +		/*
>> +		 * XXX: For now, visit_groups_merge() gets called with pmu
>> +		 * pointer never NULL. But these functions needs to be called
>> +		 * once for each pmu if I implement pmu=NULL optimization.
>> +		 */
>> +		__link_epc((*evt)->pmu_ctx);
>> +		perf_assert_pmu_disabled((*evt)->pmu_ctx->pmu);
>> +	}
>> +
>> +
>>  	min_heapify_all(&event_heap, &perf_min_heap);
>>  
>>  	while (event_heap.nr) {
> 
>> @@ -3741,39 +3799,67 @@ static int merge_sched_in(struct perf_ev
>>  	return 0;
>>  }
>>  
>> -static void
>> -ctx_pinned_sched_in(struct perf_event_context *ctx,
>> -		    struct perf_cpu_context *cpuctx)
>> +static void ctx_pinned_sched_in(struct perf_event_context *ctx, struct pmu *pmu)
>>  {
>> +	struct perf_event_pmu_context *pmu_ctx;
>>  	int can_add_hw = 1;
>>  
>> -	if (ctx != &cpuctx->ctx)
>> -		cpuctx = NULL;
>> -
>> -	visit_groups_merge(cpuctx, &ctx->pinned_groups,
>> -			   smp_processor_id(),
>> -			   merge_sched_in, &can_add_hw);
>> +	if (pmu) {
>> +		visit_groups_merge(ctx, &ctx->pinned_groups,
>> +				   smp_processor_id(), pmu,
>> +				   merge_sched_in, &can_add_hw);
>> +	} else {
>> +		/*
>> +		 * XXX: This can be optimized for per-task context by calling
>> +		 * visit_groups_merge() only once with:
>> +		 *   1) pmu=NULL
>> +		 *   2) Ignoring pmu in perf_event_groups_cmp() when it's NULL
>> +		 *   3) Making can_add_hw a per-pmu variable
>> +		 *
>> +		 * Though, it can not be opimized for per-cpu context because
>> +		 * per-cpu rb-tree consist of pmu-subtrees and pmu-subtrees
>> +		 * consist of cgroup-subtrees. i.e. a cgroup events of same
>> +		 * cgroup but different pmus are seperated out into respective
>> +		 * pmu-subtrees.
>> +		 */
>> +		list_for_each_entry(pmu_ctx, &ctx->pmu_ctx_list, pmu_ctx_entry) {
>> +			can_add_hw = 1;
>> +			visit_groups_merge(ctx, &ctx->pinned_groups,
>> +					   smp_processor_id(), pmu_ctx->pmu,
>> +					   merge_sched_in, &can_add_hw);
>> +		}
>> +	}
>>  }
> 
> I'm not sure I follow.. task context can have multiple PMUs just the
> same as CPU context can, that's more or less the entire point of the
> patch.

Current rbtree key is {cpu, cgroup_id, group_idx}. However, effective key for
task specific context is {cpu, group_idx} because cgroup_id is always 0. And
effective key for cpu specific context is {cgroup_id, group_idx} because cpu
is same for entire rbtree.

With New design, rbtree key will be {cpu, pmu, cgroup_id, group_idx}. But as
explained above, effective key for task specific context will be {cpu, pmu,
group_idx}. Thus, we can handle pmu=NULL in visit_groups_merge(), same as you
did in the very first RFC[1]. (This may make things more complicated though
because we might also need to increase heap size to accommodate all pmu events
in single heap. Current heap size is 2 for task specific context, which is
sufficient if we iterate over all pmus).

Same optimization won't work for cpu specific context because, it's effective
key would be {pmu, cgroup_id, group_idx} i.e. each pmu subtree is made up of
cgroup subtrees.

Please correct me if my understanding is wrong.

Thanks,
Ravi

[1]:
https://lore.kernel.org/lkml/20181010104559.GO5728@hirez.programming.kicks-ass.net