lkml.org 
[lkml]   [2020]   [Jul]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH V7 08/14] perf/x86/intel: Generic support for hardware TopDown metrics
From
Date


On 7/28/2020 9:09 AM, Peter Zijlstra wrote:
> On Fri, Jul 24, 2020 at 03:10:52PM -0400, Liang, Kan wrote:
>
>> diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
>> index 6cb079e0c9d9..010ac74afc09 100644
>> --- a/arch/x86/events/intel/core.c
>> +++ b/arch/x86/events/intel/core.c
>> @@ -2405,27 +2405,18 @@ static u64 icl_update_topdown_event(struct
>> perf_event *event)
>> return slots;
>> }
>>
>> -static void intel_pmu_read_topdown_event(struct perf_event *event)
>> +static void intel_pmu_read_event(struct perf_event *event)
>> {
>> - struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
>> -
>> - /* Only need to call update_topdown_event() once for group read. */
>> - if ((cpuc->txn_flags & PERF_PMU_TXN_READ) &&
>> - !is_slots_event(event))
>> return;
>>
>> - perf_pmu_disable(event->pmu);
>> - x86_pmu.update_topdown_event(event);
>> - perf_pmu_enable(event->pmu);
>> -}
>> -
>> -static void intel_pmu_read_event(struct perf_event *event)
>> -{
>> if (event->hw.flags & PERF_X86_EVENT_AUTO_RELOAD)
>> intel_pmu_auto_reload_read(event);
>> - else if (is_topdown_count(event) && x86_pmu.update_topdown_event)
>> - intel_pmu_read_topdown_event(event);
>> - else
>> + else if (is_slots_count(event) && x86_pmu.update_topdown_event) {
>> + perf_pmu_disable(event->pmu);
>> + x86_pmu.update_topdown_event(event);
>> + perf_pmu_enable(event->pmu);
>> + } else
>> x86_perf_event_update(event);
>> }
>
> I'm a little puzzled by this; what happens if you:
>
> fd = sys_perf_event_open(&attr_slots);
> fd1 = sys_perf_event_open(&attr_metric, .group_fd=fd);
>
> read(fd1);
>
> ?
>

I did a quick test. It depends on the .read_format of attr_metric.
If PERF_FORMAT_GROUP is applied for attr_metric, perf_read_group() will
be invoked. The value of fd1 is updated correctly.
If the flag is not applied, 0 will be returned.

static ssize_t
__perf_read(struct perf_event *event, char __user *buf, size_t count)
{
u64 read_format = event->attr.read_format;
int ret;

/*
* Return end-of-file for a read on an event that is in
* error state (i.e. because it was pinned but it couldn't be
* scheduled on to the CPU at some point).
*/
if (event->state == PERF_EVENT_STATE_ERROR)
return 0;

if (count < event->read_size)
return -ENOSPC;

WARN_ON_ONCE(event->ctx->parent_ctx);
if (read_format & PERF_FORMAT_GROUP)
ret = perf_read_group(event, read_format, buf);
else
ret = perf_read_one(event, read_format, buf);

return ret;
}


Thanks,
Kan

\
 
 \ /
  Last update: 2020-07-28 15:29    [W:0.092 / U:0.292 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site