Skip to contentSkip to conversation

Use case for PERF_COUNT_HW_REF_CPU_CYCLES generic PMU event

DISCUSSION Anshuman Khandual 2 participants 4 messages Feb 8, 2012

lore
Share & tools
lore
#0 lore raw

Hello Stephane,

I was going through the following discussion where we added the new HW generic event PERF_COUNT_HW_REF_CPU_CYCLES.

https://lkml.org/lkml/2011/12/10/103

(Sorry, for asking this question bit late)

I am trying to understand the use case for this. Would this new event help us in generating (during a perf session) a CPU frequency invariant time metric against which we would plot our other perf event's measurements ? CPU frequency independent time measurement is it's primary purpose ? or we were finding a way to expose the fixed counter 2 which was not getting used before for not having an event encoding. I guess this would help us in finding equivalent PMU events or mechanisms in other architecture / platforms.

Signature · 3 lines
Anshuman Khandual
Linux Technology Centre
IBM Systems and Technology Group
#1 lore raw
Quoted 9 lines · Anshuman Khandual

Hello Stephane,

I was going through the following discussion where we added the new HW generic event PERF_COUNT_HW_REF_CPU_CYCLES.

https://lkml.org/lkml/2011/12/10/103

(Sorry, for asking this question bit late)

Quoted 8 lines

I am trying to understand the use case for this. Would this new event help us in generating (during a perf session) a CPU frequency invariant time metric against which we would plot our other perf event's measurements ? CPU frequency independent time measurement is it's primary purpose ? or we were finding a way to expose the fixed counter 2 which was not getting used before for not having an event encoding. I guess this would help us in finding equivalent PMU events or mechanisms in other architecture / platforms.

The goal was to expose a cycle event that is not subject to frequency scaling nor turbo boost of any sort. An event that could be used to correlate with time. An event that could also be used to compute idle time by comparing its value with wall-clock time.

The fact that on Intel X86 this event is on fixed counter 2 is an implementation
detail.

Quoted 5 lines

--
Anshuman Khandual
Linux Technology Centre
IBM Systems and Technology Group

#2 lore raw
Quoted 19 lines · Stephane Eranian
Quoted 9 lines · Anshuman Khandual

Hello Stephane,

I was going through the following discussion where we added the new HW generic event PERF_COUNT_HW_REF_CPU_CYCLES.

https://lkml.org/lkml/2011/12/10/103

(Sorry, for asking this question bit late)

Quoted 8 lines

I am trying to understand the use case for this. Would this new event help us in generating (during a perf session) a CPU frequency invariant time metric against which we would plot our other perf event's measurements ? CPU frequency independent time measurement is it's primary purpose ? or we were finding a way to expose the fixed counter 2 which was not getting used before for not having an event encoding. I guess this would help us in finding equivalent PMU events or mechanisms in other architecture / platforms.

Quoted 5 lines

The goal was to expose a cycle event that is not subject to frequency scaling nor turbo boost of any sort. An event that could be used to correlate with time.

An event that could also be used to compute idle time by comparing its value with wall-clock time.

Why kernel computed idle time is not sufficient ? How much accuracy would it improve in using PMU event computed idle time over kernel computed idle time.

Quoted 11 lines

The fact that on Intel X86 this event is on fixed counter 2 is an implementation
detail.

Quoted 5 lines

--
Anshuman Khandual
Linux Technology Centre
IBM Systems and Technology Group

#3 lore raw
Quoted 20 lines · Anshuman Khandual
Quoted 19 lines · Stephane Eranian
Quoted 17 lines · Anshuman Khandual

Hello Stephane,

I was going through the following discussion where we added the new HW generic event PERF_COUNT_HW_REF_CPU_CYCLES.

https://lkml.org/lkml/2011/12/10/103

(Sorry, for asking this question bit late)

I am trying to understand the use case for this. Would this new event help us in generating (during a perf session) a CPU frequency invariant time metric against which we would plot our other perf event's measurements ? CPU frequency independent time measurement is it's primary purpose ? or we were finding a way to expose the fixed counter 2 which was not getting used before for not having an event encoding. I guess this would help us in finding equivalent PMU events or mechanisms in other architecture / platforms.

Quoted 9 lines
Quoted 5 lines

The goal was to expose a cycle event that is not subject to frequency scaling nor turbo boost of any sort. An event that could be used to correlate with time.

An event that could also be used to compute idle time by comparing its value with wall-clock time.

Why kernel computed idle time is not sufficient ? How much accuracy would it improve in using PMU event computed idle time over kernel computed idle time.

I was just listing the what you could do with the event. I have not yet tried to measure
idle with ref-cycles.

Quoted 13 lines
Quoted 12 lines

The fact that on Intel X86 this event is on fixed counter 2 is an implementation
detail.

Quoted 5 lines

--
Anshuman Khandual
Linux Technology Centre
IBM Systems and Technology Group