AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[SWDEV-463406] Update API with fields for gfx_clock_below_host_limit and low_utilization violations

Browse Source 
Updated API with fields for gfx_clock_below_host_limit and low_utilization violations
Change-Id: I25647bae6e7b785f44dab024272767658688bcad

---------
Signed-off-by: Scaffidi, Salvatore <Salvatore.Scaffidi@amd.com>
Signed-off-by: Arif, Maisam <Maisam.Arif@amd.com>
Co-authored-by: Charis Poag <Charis.Poag@amd.com>
This commit is contained in:
Scaffidi, Salvatore
2025-01-08 22:07:23 -06:00
committed by GitHub
parent 490132748f
commit 3793be7735
8 changed files with 245 additions and 84 deletions
Download Patch File Download Diff File Expand all files Collapse all files
py-interface/amdsmi_interface.py
+10 -3
View File
@@ -28,6 +28,7 @@ from .amdsmi_exception import *
import sys
import math
from time import localtime, asctime, time
import json
MAX_NUM_PROCESSES = 1024
@@ -1559,7 +1560,9 @@ def amdsmi_get_hsmp_metrics_table(
"mtbl_ppt_residency_acc": mtbl.ppt_residency_acc,
"mtbl_socket_thm_residency_acc": mtbl.socket_thm_residency_acc,
"mtbl_vr_thm_residency_acc": mtbl.vr_thm_residency_acc,
"mtbl_hbm_thm_residency_acc": mtbl.hbm_thm_residency_acc
"mtbl_hbm_thm_residency_acc": mtbl.hbm_thm_residency_acc,
"mtbl_gfx_clk_below_host_residency_acc": mtbl.gfx_clk_below_host_residency_acc,
"mtbl_low_utilization_residency_acc": mtbl.low_utilization_residency_acc
}
def amdsmi_first_online_core_on_cpu_socket(
@@ -2035,7 +2038,7 @@ def amdsmi_get_violation_status(
processor_handle, ctypes.byref(violation_status))
)
return {
dict_return = {
"reference_timestamp": _validate_if_max_uint(violation_status.reference_timestamp, MaxUIntegerTypes.UINT64_T),
"violation_timestamp": _validate_if_max_uint(violation_status.violation_timestamp, MaxUIntegerTypes.UINT64_T),
"acc_counter": _validate_if_max_uint(violation_status.acc_counter, MaxUIntegerTypes.UINT64_T),
@@ -2044,17 +2047,21 @@ def amdsmi_get_violation_status(
"acc_socket_thrm": _validate_if_max_uint(violation_status.acc_socket_thrm, MaxUIntegerTypes.UINT64_T), #TVIOL
"acc_vr_thrm": _validate_if_max_uint(violation_status.acc_vr_thrm, MaxUIntegerTypes.UINT64_T),
"acc_hbm_thrm": _validate_if_max_uint(violation_status.acc_hbm_thrm, MaxUIntegerTypes.UINT64_T),
"acc_gfx_clk_below_host_limit": _validate_if_max_uint(violation_status.acc_gfx_clk_below_host_limit, MaxUIntegerTypes.UINT64_T),
"per_prochot_thrm": _validate_if_max_uint(violation_status.per_prochot_thrm, MaxUIntegerTypes.UINT64_T, isActivity=True),
"per_ppt_pwr": _validate_if_max_uint(violation_status.per_ppt_pwr, MaxUIntegerTypes.UINT64_T, isActivity=True), #PVIOL
"per_socket_thrm": _validate_if_max_uint(violation_status.per_socket_thrm, MaxUIntegerTypes.UINT64_T, isActivity=True), #TVIOL
"per_vr_thrm": _validate_if_max_uint(violation_status.per_vr_thrm, MaxUIntegerTypes.UINT64_T, isActivity=True),
"per_hbm_thrm": _validate_if_max_uint(violation_status.per_hbm_thrm, MaxUIntegerTypes.UINT64_T, isActivity=True),
"per_gfx_clk_below_host_limit": _validate_if_max_uint(violation_status.per_gfx_clk_below_host_limit, MaxUIntegerTypes.UINT64_T, isActivity=True),
"active_prochot_thrm": _validate_if_max_uint(violation_status.active_prochot_thrm, MaxUIntegerTypes.UINT8_T, isBool=True),
"active_ppt_pwr": _validate_if_max_uint(violation_status.active_ppt_pwr, MaxUIntegerTypes.UINT8_T, isBool=True), #PVIOL
"active_socket_thrm": _validate_if_max_uint(violation_status.active_socket_thrm, MaxUIntegerTypes.UINT8_T, isBool=True), #TVIOL
"active_vr_thrm": _validate_if_max_uint(violation_status.active_vr_thrm, MaxUIntegerTypes.UINT8_T, isBool=True),
"active_hbm_thrm": _validate_if_max_uint(violation_status.active_hbm_thrm, MaxUIntegerTypes.UINT8_T, isBool=True)
"active_hbm_thrm": _validate_if_max_uint(violation_status.active_hbm_thrm, MaxUIntegerTypes.UINT8_T, isBool=True),
"active_gfx_clk_below_host_limit": _validate_if_max_uint(violation_status.active_gfx_clk_below_host_limit, MaxUIntegerTypes.UINT8_T, isBool=True),
}
return dict_return
def amdsmi_get_gpu_total_ecc_count(
processor_handle: amdsmi_wrapper.amdsmi_processor_handle,