add violation metrics 1.8 version changes adding partition level violation stats

[spraveenio]

compilation successful

-labsl_cordz_functions -labsl_exponential_biased -labsl_status -labsl_statusor -labsl_flags_commandlineflag_internal -labsl_flags_commandlineflag -labsl_flags_private_handle_accesso
r -labsl_flags_marshalling -labsl_flags_program_name -labsl_flags_config -labsl_flags_internal -labsl_flags_reflection -labsl_flags -labsl_flags_usage_internal -labsl_flags_usage -l
absl_flags_parse -lpthread -lz -lm -lrt -ldl -l:libev.a -l:libzmq.a -l:libssl.a -l:libcrypto.a -o /usr/src/github.com/ROCm/gpu-agent/sw/nic/build/x86_64/sim/bin/gpuagent_mock
make gpuctl
make[1]: Entering directory '/usr/src/github.com/ROCm/gpu-agent/sw/nic/gpuagent'
building gpuctl
CGO_ENABLED=0 go build -C cli -o /usr/src/github.com/ROCm/gpu-agent/sw/nic/build/x86_64/sim/bin/gpuctl

hw gpu without violation enabled

[root@exporter-amdgpu-metrics-exporter-bjmpj ~]# gpuctl show gpu all
 -- snipped ---
  VRAM usage:
    Total VRAM (in MB)                   : 196592
    Used VRAM (in MB)                    : 1525
    Free VRAM (in MB)                    : 195067
    Total visible VRAM (in MB)           : 196592
    Used visible VRAM (in MB)            : 1525
    Free visible VRAM (in MB)            : 195067
    Total GTT (in MB)                    : 773918
    Used GTT (in MB)                     : 26
    Free GTT (in MB)                     : 773892
  Accumulated energy consumed (in uJ)    : 445856595574784.00
  GFX activity accumulated               : 189101231
  Memory activity accumulated            : 4132638
  Link 2 data read (in KB)               : 1
  Link 2 data written (in KB)            : 1
  Link 3 data read (in KB)               : 1
  Link 3 data written (in KB)            : 1
  Link 4 data read (in KB)               : 1
  Link 4 data written (in KB)            : 1
  Link 5 data read (in KB)               : 1
  Link 5 data written (in KB)            : 1
  Link 6 data read (in KB)               : 1
  Link 6 data written (in KB)            : 1
  Link 7 data read (in KB)               : 1
  Link 7 data written (in KB)            : 1
  Link 8 data read (in KB)               : 1
  Link 8 data written (in KB)            : 1
  Current accumulated counter            : 2467541803
  Processor hot residency accumulated    : 0
  PPT residency accumulated              : 297918
  Socket thermal residency accumulated   : 3295
  VR thermal residency accumulated       : 0
  HBM thermal residency accumulated      : 0
  Processor hot residency percentage     : 0%
  PPT residency percentage               : 0%
  Socket thermal residency percentage    : 0%
  VR thermal residency percentage        : 0%
  HBM thermal residency percentage       : 0%

[root@exporter-amdgpu-metrics-exporter-bjmpj ~]# gpuctl show gpu -y
    violationstats:
      currentaccumulatedcounter: 2468312606
      processorhotresidencyaccumulated: 0
      pptresidencyaccumulated: 297918
      socketthermalresidencyaccumulated: 3295
      vrthermalresidencyaccumulated: 0
      hbmthermalresidencyaccumulated: 0
      processorhotresidencypercentage: 0
      pptresidencypercentage: 0
      socketthermalresidencypercentage: 0
      vrthermalresidencypercentage: 0
      hbmthermalresidencypercentage: 0
      gfxbelowhostlimitpoweraccumulated:
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      gfxbelowhostlimitthmaccumulated:
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      gfxlowutilizationaccumulated:
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      gfxbelowhostlimittotalaccumulated:
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      gfxbelowhostlimitpowerpercentage:
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      gfxbelowhostlimitthmpercentage:
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      gfxlowutilizationpercentage:
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      gfxbelowhostlimittotalpercentage:
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      - 18446744073709551615
      xxx_nounkeyedliteral: {}
      xxx_unrecognized: []
      xxx_sizecache: 0
    xxx_nounkeyedliteral: {}
    xxx_unrecognized: []
    xxx_sizecache: 0
  xxx_nounkeyedliteral: {}
  xxx_unrecognized: []
  xxx_sizecache: 0

gpu mock

[root@b15a1a105561 /]# gpuctl show gpu all
Id                                       : 0003ac11-420f-4242-0f51-d89b47c87169 (15)
--- snipped ---
  Accumulated energy consumed (in uJ)    : 25293978861511.00
  Current accumulated counter            : 123478
  Processor hot residency accumulated    : 23443
  PPT residency accumulated              : 34523
  Socket thermal residency accumulated   : 45687
  VR thermal residency accumulated       : 56753
  HBM thermal residency accumulated      : 67869
  Processor hot residency percentage     : 7%
  PPT residency percentage               : 22%
  Socket thermal residency percentage    : 41%
  VR thermal residency percentage        : 44%
  HBM thermal residency percentage       : 8%
  GPU GFX clock host limit accumulated:
    Power                                : 1225 1219 1241 1231 1171 1255 1221 1235
    Thermal                              : 2285 2299 2348 2368 2430 2403 2285 2393
    Low Utilization                      : 3501 3426 3465 3414 3409 3516 3442 3446
    Total                                : 4561 4620 4609 4535 4599 4599 4602 4565
  GPU GFX clock host limit percentage:
    Power                                : 19% 90% 10% 90% 43% 17% 46% 54%
    Thermal                              : 65% 27% 20% 48% 64% 34% 28% 36%
    Low Utilization                      : 27% 50% 53% 1% 65% 65% 82% 12%
    Total                                : 59% 20% 34% 68% 39% 49% 35% 62%

(cherry picked from commit 788653d006e52ea9513972792222e4b4ac37d00a)

Motivation

Technical Details

Test Plan

Test Result

Submission Checklist

• Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

[sarat-k]

LGTM