Merge amd-staging into amd-master 20230302

Signed-off-by: Hao Zhou <Hao.Zhou@amd.com>
Change-Id: I17c723b907c988754d2da5d04092c1d435e29cc1

python_smi_tools/rocm_smi.py

@@ -1440,6 +1440,9 @@ def setComputePartition(deviceList, computePartitionType):
                 None)
         elif ret == rsmi_status_t.RSMI_STATUS_PERMISSION:
             printLog(device, 'Permission denied', None)
+        elif ret == rsmi_status_t.RSMI_STATUS_SETTING_UNAVAILABLE:
+            printLog(device, 'Requested setting (%s) is unavailable for current device'
+                     %computePartitionType, None)
         elif ret == rsmi_status_t.RSMI_STATUS_NOT_SUPPORTED:
             printLog(device, 'Not supported on the given system', None)
         else:
@@ -1558,13 +1561,12 @@ def showAllConcise(deviceList):
         else:
             gpu_busy = 'Unsupported'
         memInfo = getMemInfo(device, 'vram')
-        if memInfo[0] == None or memInfo[1] == None:
-            mem_use = 'N/A'
-        else:
+        mem_use_pct = 0
+        if memInfo[0] != None and memInfo[1] != None and float(memInfo[1]) != 0:
             mem_use_pct = '% 3.0f%%' % (100 * (float(memInfo[0]) / float(memInfo[1])))
         values['card%s' % (str(device))] = [device, temp, avgPwr, sclk, mclk, fan, str(perf).lower(), pwrCap,
                                             mem_use_pct, gpu_busy]
-        val_widths = {}
+    val_widths = {}
     for device in deviceList:
         val_widths[device] = [len(str(val)) + 2 for val in values['card%s' % (str(device))]]
     max_widths = head_widths

python_smi_tools/rsmiBindings.py

@@ -69,7 +69,8 @@ class rsmi_status_t(c_int):
     RSMI_STATUS_UNEXPECTED_DATA = 0xF
     RSMI_STATUS_BUSY = 0x10
     RSMI_STATUS_REFCOUNT_OVERFLOW = 0x11
-    RSMI_STATUS_AMDGPU_RESTART_ERR = 0x12
+    RSMI_STATUS_SETTING_UNAVAILABLE = 0x12
+    RSMI_STATUS_AMDGPU_RESTART_ERR = 0x13
     RSMI_STATUS_UNKNOWN_ERROR = 0xFFFFFFFF
 
 
@@ -93,6 +94,7 @@ rsmi_status_verbose_err_out = {
     rsmi_status_t.RSMI_STATUS_UNEXPECTED_DATA: 'Unexpected data received',
     rsmi_status_t.RSMI_STATUS_BUSY: 'Busy - resources are preventing call the ability to execute',
     rsmi_status_t.RSMI_STATUS_REFCOUNT_OVERFLOW: 'Data overflow - data exceeded INT32_MAX',
+    rsmi_status_t.RSMI_STATUS_SETTING_UNAVAILABLE: 'Requested setting is unavailable for current device',
     rsmi_status_t.RSMI_STATUS_AMDGPU_RESTART_ERR: 'Could not successfully restart the amdgpu driver',
     rsmi_status_t.RSMI_STATUS_UNKNOWN_ERROR: 'Unknown error occured'
 }