Re: [PATCH 0/6] Add per-core RAPL energy counter support for AMD CPUs

[Oleksandr Natalenko <oleksandr@natalenko.name>]

[-- Attachment #1: Type: text/plain, Size: 5365 bytes --]

On pondělí 10. června 2024 17:17:42, SELČ Dhananjay Ugwekar wrote:
> Hello Oleksandr,
> 
> On 6/10/2024 7:58 PM, Oleksandr Natalenko wrote:
> > Hello.
> > 
> > On pondělí 10. června 2024 12:07:45, SELČ Dhananjay Ugwekar wrote:
> >> Currently the energy-cores event in the power PMU aggregates energy
> >> consumption data at a package level. On the other hand the core energy
> >> RAPL counter in AMD CPUs has a core scope (which means the energy 
> >> consumption is recorded separately for each core). Earlier efforts to add
> >> the core event in the power PMU had failed [1], due to the difference in 
> >> the scope of these two events. Hence, there is a need for a new core scope
> >> PMU.
> >>
> >> This patchset adds a new "power_per_core" PMU alongside the existing
> >> "power" PMU, which will be responsible for collecting the new
> >> "energy-per-core" event.
> >>
> >> Tested the package level and core level PMU counters with workloads
> >> pinned to different CPUs.
> >>
> >> Results with workload pinned to CPU 1 in Core 1 on an AMD Zen4 Genoa 
> >> machine:
> >>
> >> $ perf stat -a --per-core -e power_per_core/energy-per-core/ sleep 1
> >>
> >>  Performance counter stats for 'system wide':
> >>
> >> S0-D0-C0         1          0.02 Joules power_per_core/energy-per-core/
> >> S0-D0-C1         1          5.72 Joules power_per_core/energy-per-core/
> >> S0-D0-C2         1          0.02 Joules power_per_core/energy-per-core/
> >> S0-D0-C3         1          0.02 Joules power_per_core/energy-per-core/
> >> S0-D0-C4         1          0.02 Joules power_per_core/energy-per-core/
> >> S0-D0-C5         1          0.02 Joules power_per_core/energy-per-core/
> >> S0-D0-C6         1          0.02 Joules power_per_core/energy-per-core/
> >> S0-D0-C7         1          0.02 Joules power_per_core/energy-per-core/
> >> S0-D0-C8         1          0.02 Joules power_per_core/energy-per-core/
> >> S0-D0-C9         1          0.02 Joules power_per_core/energy-per-core/
> >> S0-D0-C10        1          0.02 Joules power_per_core/energy-per-core/
> >>
> >> [1]: https://lore.kernel.org/lkml/3e766f0e-37d4-0f82-3868-31b14228868d@linux.intel.com/
> >>
> >> This patchset applies cleanly on top of v6.10-rc3 as well as latest 
> >> tip/master.
> >>
> >> Dhananjay Ugwekar (6):
> >>   perf/x86/rapl: Fix the energy-pkg event for AMD CPUs
> >>   perf/x86/rapl: Rename rapl_pmu variables
> >>   perf/x86/rapl: Make rapl_model struct global
> >>   perf/x86/rapl: Move cpumask variable to rapl_pmus struct
> >>   perf/x86/rapl: Add wrapper for online/offline functions
> >>   perf/x86/rapl: Add per-core energy counter support for AMD CPUs
> >>
> >>  arch/x86/events/rapl.c | 311 ++++++++++++++++++++++++++++++-----------
> >>  1 file changed, 233 insertions(+), 78 deletions(-)
> >>
> >>
> > 
> > With my CPU:
> > 
> >   Model name:             AMD Ryzen 9 5950X 16-Core Processor
> > 
> > and this workload:
> > 
> > $ taskset -c 1 dd if=/dev/zero of=/dev/null
> > 
> > the following result is got:
> > 
> > $ sudo perf stat -a --per-core -e power_per_core/energy-per-core/ sleep 1
> > 
> >  Performance counter stats for 'system wide':
> > 
> > S0-D0-C0              1               1,70 Joules power_per_core/energy-per-core/
> > S0-D0-C1              1               8,83 Joules power_per_core/energy-per-core/
> > S0-D0-C2              1               0,17 Joules power_per_core/energy-per-core/
> > S0-D0-C3              1               0,33 Joules power_per_core/energy-per-core/
> > S0-D0-C4              1               0,14 Joules power_per_core/energy-per-core/
> > S0-D0-C5              1               0,33 Joules power_per_core/energy-per-core/
> > S0-D0-C6              1               0,25 Joules power_per_core/energy-per-core/
> > S0-D0-C7              1               0,19 Joules power_per_core/energy-per-core/
> > S0-D0-C8              1               0,66 Joules power_per_core/energy-per-core/
> > S0-D0-C9              1               1,71 Joules power_per_core/energy-per-core/
> > S0-D0-C10             1               0,38 Joules power_per_core/energy-per-core/
> > S0-D0-C11             1               1,69 Joules power_per_core/energy-per-core/
> > S0-D0-C12             1               0,22 Joules power_per_core/energy-per-core/
> > S0-D0-C13             1               0,11 Joules power_per_core/energy-per-core/
> > S0-D0-C14             1               0,49 Joules power_per_core/energy-per-core/
> > S0-D0-C15             1               0,37 Joules power_per_core/energy-per-core/
> > 
> >        1,002409590 seconds time elapsed
> > 
> > If it is as expected, please add my:
> > 
> > Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> 
> We can see that after you affined the workload to cpu 1, energy 
> consumption of core 1 is considerably higher than the other cores, 
> which is as expected, will add your tested-by in next version.
> 
> P.S: I'm assuming here that cpu 1 is part of core 1 in your system, 
> please let me know if that assumption is wrong.

You assumption should be correct:

$ cat /sys/devices/system/cpu/cpu1/topology/core_id
1

> Thanks for testing the patch!
> 
> Regards,
> Dhananjay
> 
> > 
> > Thank you.
> > 
> 


-- 
Oleksandr Natalenko (post-factum)

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 833 bytes --]