Re: [PATCH v5 0/9] Add per-core RAPL energy counter support for AMD CPUs

[Oleksandr Natalenko]

Hello.

On pátek 13. září 2024 17:21:40, 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/
> 
> v4 Link: https://lore.kernel.org/all/20240711102436.4432-1-Dhananjay.Ugwekar@xxxxxxx/
> 
> v5 changes:
> * Rebase on top of Kan Liang's "PMU scope" patchset [2]
> * rapl_cntr_mask moved to rapl_pmus struct in patch 8
> * Patch 1 from v4 is merged separately, so removed from this series
> * Add an extra argument "scope" in patch 5 to the init functions
> * Add an new patch 2, which removes the cpu_to_rapl_pmu() function
> 
> Base: tip/perf/core(currently has just 1-5 patches from [2]) + patch 6 from [2] + 
>       diff [3] + patch 7 from [2] + revert [4] + apply [5] 
> 
> [1]: https://lore.kernel.org/lkml/3e766f0e-37d4-0f82-3868-31b14228868d@xxxxxxxxxxxxxxx/
> [2]: https://lore.kernel.org/all/20240802151643.1691631-1-kan.liang@xxxxxxxxxxxxxxx/
> [3]: https://lore.kernel.org/all/8c09633c-5bf2-48a2-91a6-a0af9b9f2e8c@xxxxxxxxxxxxxxx/
> [4]: https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?h=perf/core&id=8d72eba1cf8cecd76a2b4c1dd7673c2dc775f514
> [5]: https://lore.kernel.org/all/20240910085504.204814-1-Dhananjay.Ugwekar@xxxxxxx/
> 
> Dhananjay Ugwekar (8):
>   perf/x86/rapl: Remove the cpu_to_rapl_pmu() function
>   perf/x86/rapl: Rename rapl_pmu variables
>   perf/x86/rapl: Make rapl_model struct global
>   perf/x86/rapl: Add arguments to the cleanup and init functions
>   perf/x86/rapl: Modify the generic variable names to *_pkg*
>   perf/x86/rapl: Remove the global variable rapl_msrs
>   perf/x86/rapl: Move the cntr_mask to rapl_pmus struct
>   perf/x86/rapl: Add per-core energy counter support for AMD CPUs
> 
> K Prateek Nayak (1):
>   x86/topology: Introduce topology_logical_core_id()
> 
>  Documentation/arch/x86/topology.rst   |   4 +
>  arch/x86/events/rapl.c                | 406 ++++++++++++++++----------
>  arch/x86/include/asm/processor.h      |   1 +
>  arch/x86/include/asm/topology.h       |   1 +
>  arch/x86/kernel/cpu/debugfs.c         |   1 +
>  arch/x86/kernel/cpu/topology_common.c |   1 +
>  6 files changed, 266 insertions(+), 148 deletions(-)

With v6.11-rc7 + all the mentioned preparatory patches and this series:

$ taskset -c 9 dd if=/dev/zero of=/dev/null &

$ sudo perf stat -a --per-core -e power_per_core/energy-per-core/ sleep 5

 Performance counter stats for 'system wide':

S0-D0-C0              1               3,79 Joules power_per_core/energy-per-core/
S0-D0-C1              1               5,65 Joules power_per_core/energy-per-core/
S0-D0-C2              1               1,26 Joules power_per_core/energy-per-core/
S0-D0-C3              1               3,18 Joules power_per_core/energy-per-core/
S0-D0-C4              1               2,06 Joules power_per_core/energy-per-core/
S0-D0-C5              1               3,51 Joules power_per_core/energy-per-core/
S0-D0-C6              1               0,77 Joules power_per_core/energy-per-core/
S0-D0-C7              1               0,55 Joules power_per_core/energy-per-core/
S0-D0-C8              1               1,65 Joules power_per_core/energy-per-core/
S0-D0-C9              1              47,85 Joules power_per_core/energy-per-core/
S0-D0-C10             1               2,49 Joules power_per_core/energy-per-core/
S0-D0-C11             1              11,85 Joules power_per_core/energy-per-core/
S0-D0-C12             1               1,75 Joules power_per_core/energy-per-core/
S0-D0-C13             1               0,74 Joules power_per_core/energy-per-core/
S0-D0-C14             1               2,58 Joules power_per_core/energy-per-core/
S0-D0-C15             1               4,67 Joules power_per_core/energy-per-core/

       5,003391425 seconds time elapsed

on the following CPU:

AMD Ryzen 9 5950X 16-Core Processor

If this behaves as expected, please add:

Tested-by: Oleksandr Natalenko <oleksandr@xxxxxxxxxxxxxx>

Thank you.

-- 
Oleksandr Natalenko (post-factum)

Attachment: signature.asc
Description: This is a digitally signed message part.