Browse Source

nvml: clean up and fix retorted bus id bug in api

2upstream
Tanguy Pruvot 10 years ago
parent
commit
7bb86defcc
  1. 144
      nvml.cpp
  2. 6
      nvml.h

144
nvml.cpp

@ -200,14 +200,14 @@ wrap_nvml_handle * wrap_nvml_create() @@ -200,14 +200,14 @@ wrap_nvml_handle * wrap_nvml_create()
nvmlh->cuda_nvml_device_id[i] = -1;
if (cudaGetDeviceProperties(&props, i) == cudaSuccess) {
int j;
device_bus_ids[i] = props.pciBusID;
for (j = 0; j<nvmlh->nvml_gpucount; j++) {
for (int j = 0; j < nvmlh->nvml_gpucount; j++) {
if ((nvmlh->nvml_pci_domain_id[j] == (uint32_t) props.pciDomainID) &&
(nvmlh->nvml_pci_bus_id[j] == (uint32_t) props.pciBusID) &&
(nvmlh->nvml_pci_device_id[j] == (uint32_t) props.pciDeviceID)) {
if (opt_debug)
applog(LOG_DEBUG, "CUDA GPU[%d] matches NVML GPU[%d]", i, j);
applog(LOG_DEBUG, "CUDA GPU#%d matches NVML GPU %d by busId %u",
i, j, (uint32_t) props.pciBusID);
nvmlh->nvml_cuda_device_id[j] = i;
nvmlh->cuda_nvml_device_id[i] = j;
}
@ -275,24 +275,6 @@ int wrap_nvml_get_fanpcnt(wrap_nvml_handle *nvmlh, int cudaindex, unsigned int * @@ -275,24 +275,6 @@ int wrap_nvml_get_fanpcnt(wrap_nvml_handle *nvmlh, int cudaindex, unsigned int *
return 0;
}
/* Not Supported on 750Ti 340.23, 346.16 neither */
int wrap_nvml_get_clock(wrap_nvml_handle *nvmlh, int cudaindex, int type, unsigned int *freq)
{
int gpuindex = nvmlh->cuda_nvml_device_id[cudaindex];
if (gpuindex < 0 || gpuindex >= nvmlh->nvml_gpucount)
return -1;
// wrap_nvmlReturn_t rc = nvmlh->nvmlDeviceGetApplicationsClock(nvmlh->devs[gpuindex], (wrap_nvmlClockType_t)type, freq);
wrap_nvmlReturn_t rc = nvmlh->nvmlDeviceGetClockInfo(nvmlh->devs[gpuindex], (wrap_nvmlClockType_t) type, freq);
if (rc != WRAPNVML_SUCCESS) {
//if (opt_debug)
// applog(LOG_DEBUG, "nvmlDeviceGetClockInfo: %s", nvmlh->nvmlErrorString(rc));
return -1;
}
return 0;
}
/* Not Supported on 750Ti 340.23 */
int wrap_nvml_get_power_usage(wrap_nvml_handle *nvmlh, int cudaindex, unsigned int *milliwatts)
{
@ -425,31 +407,6 @@ int nvapi_fanspeed(unsigned int devNum, unsigned int *speed) @@ -425,31 +407,6 @@ int nvapi_fanspeed(unsigned int devNum, unsigned int *speed)
return 0;
}
int nvapi_getclock(unsigned int devNum, unsigned int *freq)
{
NvAPI_Status ret;
if (devNum >= nvapi_dev_cnt)
return -1;
NV_GPU_CLOCK_FREQUENCIES_V2 clocks;
clocks.version = NV_GPU_CLOCK_FREQUENCIES_VER_2;
clocks.ClockType = NV_GPU_CLOCK_FREQUENCIES_CURRENT_FREQ; // CURRENT/BASE/BOOST
ret = NvAPI_GPU_GetAllClockFrequencies(phys[devNum], &clocks);
if (ret != NVAPI_OK) {
NvAPI_ShortString string;
NvAPI_GetErrorMessage(ret, string);
if (opt_debug)
applog(LOG_DEBUG, "NVAPI NvAPI_GPU_GetAllClockFrequencies: %s", string);
return -1;
} else {
// GRAPHICS/MEMORY
(*freq) = (unsigned int)clocks.domain[NVAPI_GPU_PUBLIC_CLOCK_GRAPHICS].frequency;
}
return 0;
}
int nvapi_getpstate(unsigned int devNum, unsigned int *power)
{
NvAPI_Status ret;
@ -519,15 +476,6 @@ int nvapi_getinfo(unsigned int devNum, char *desc) @@ -519,15 +476,6 @@ int nvapi_getinfo(unsigned int devNum, char *desc)
return 0;
}
int nvapi_getbusid(unsigned int devNum, int *busid)
{
if (devNum >= 0 && devNum <= 8) {
(*busid) = device_bus_ids[devNum];
return 0;
}
return -1;
}
int wrap_nvapi_init()
{
int num_gpus = cuda_num_devices();
@ -567,8 +515,8 @@ int wrap_nvapi_init() @@ -567,8 +515,8 @@ int wrap_nvapi_init()
if (ret == NVAPI_OK && busId == device_bus_ids[g]) {
nvapi_dev_map[g] = i;
if (opt_debug)
applog(LOG_DEBUG, "CUDA GPU[%d] matches NVAPI GPU[%d]",
g, i);
applog(LOG_DEBUG, "CUDA GPU#%d matches NVAPI GPU %d by busId %u",
g, i, busId);
break;
}
}
@ -657,7 +605,7 @@ int gpu_busid(struct cgpu_info *gpu) @@ -657,7 +605,7 @@ int gpu_busid(struct cgpu_info *gpu)
}
#ifdef WIN32
if (support == -1) {
nvapi_getbusid(nvapi_dev_map[gpu->gpu_id], &busid);
busid = device_bus_ids[gpu->gpu_id];
}
#endif
return busid;
@ -675,6 +623,7 @@ unsigned int gpu_power(struct cgpu_info *gpu) @@ -675,6 +623,7 @@ unsigned int gpu_power(struct cgpu_info *gpu)
if (support == -1) {
unsigned int pct = 0;
nvapi_getusage(nvapi_dev_map[gpu->gpu_id], &pct);
mw = pct; // to fix
}
#endif
return mw;
@ -704,82 +653,3 @@ int gpu_clocks(struct cgpu_info *gpu) @@ -704,82 +653,3 @@ int gpu_clocks(struct cgpu_info *gpu)
}
return -1;
}
/* strings /usr/lib/nvidia-340/libnvidia-ml.so | grep nvmlDeviceGet | grep -v : | sort | uniq
nvmlDeviceGetAccountingBufferSize
nvmlDeviceGetAccountingMode
nvmlDeviceGetAccountingPids
nvmlDeviceGetAccountingStats
nvmlDeviceGetAPIRestriction
nvmlDeviceGetApplicationsClock
nvmlDeviceGetAutoBoostedClocksEnabled
nvmlDeviceGetBAR1MemoryInfo
nvmlDeviceGetBoardId
nvmlDeviceGetBrand
nvmlDeviceGetBridgeChipInfo
* nvmlDeviceGetClockInfo
nvmlDeviceGetComputeMode
nvmlDeviceGetComputeRunningProcesses
nvmlDeviceGetCount
nvmlDeviceGetCount_v2
nvmlDeviceGetCpuAffinity
nvmlDeviceGetCurrentClocksThrottleReasons
nvmlDeviceGetCurrPcieLinkGeneration
nvmlDeviceGetCurrPcieLinkWidth
nvmlDeviceGetDecoderUtilization
nvmlDeviceGetDefaultApplicationsClock
nvmlDeviceGetDetailedEccErrors
nvmlDeviceGetDisplayActive
nvmlDeviceGetDisplayMode
nvmlDeviceGetDriverModel
nvmlDeviceGetEccMode
nvmlDeviceGetEncoderUtilization
nvmlDeviceGetEnforcedPowerLimit
* nvmlDeviceGetFanSpeed
nvmlDeviceGetGpuOperationMode
nvmlDeviceGetHandleByIndex
* nvmlDeviceGetHandleByIndex_v2
nvmlDeviceGetHandleByPciBusId
nvmlDeviceGetHandleByPciBusId_v2
nvmlDeviceGetHandleBySerial
nvmlDeviceGetHandleByUUID
nvmlDeviceGetIndex
nvmlDeviceGetInforomConfigurationChecksum
nvmlDeviceGetInforomImageVersion
nvmlDeviceGetInforomVersion
nvmlDeviceGetMaxClockInfo
nvmlDeviceGetMaxPcieLinkGeneration
nvmlDeviceGetMaxPcieLinkWidth
nvmlDeviceGetMemoryErrorCounter
nvmlDeviceGetMemoryInfo
nvmlDeviceGetMinorNumber
nvmlDeviceGetMultiGpuBoard
* nvmlDeviceGetName
* nvmlDeviceGetPciInfo
nvmlDeviceGetPciInfo_v2
* nvmlDeviceGetPerformanceState
nvmlDeviceGetPersistenceMode
nvmlDeviceGetPowerManagementDefaultLimit
nvmlDeviceGetPowerManagementLimit
nvmlDeviceGetPowerManagementLimitConstraints
nvmlDeviceGetPowerManagementMode
nvmlDeviceGetPowerState (deprecated)
* nvmlDeviceGetPowerUsage
nvmlDeviceGetRetiredPages
nvmlDeviceGetRetiredPagesPendingStatus
nvmlDeviceGetSamples
nvmlDeviceGetSerial
nvmlDeviceGetSupportedClocksThrottleReasons
nvmlDeviceGetSupportedEventTypes
nvmlDeviceGetSupportedGraphicsClocks
nvmlDeviceGetSupportedMemoryClocks
nvmlDeviceGetTemperature
nvmlDeviceGetTemperatureThreshold
nvmlDeviceGetTotalEccErrors
nvmlDeviceGetUtilizationRates
nvmlDeviceGetUUID
nvmlDeviceGetVbiosVersion
nvmlDeviceGetViolationStatus
*/

6
nvml.h

@ -111,12 +111,6 @@ int wrap_nvml_get_tempC(wrap_nvml_handle *nvmlh, @@ -111,12 +111,6 @@ int wrap_nvml_get_tempC(wrap_nvml_handle *nvmlh,
int wrap_nvml_get_fanpcnt(wrap_nvml_handle *nvmlh,
int gpuindex, unsigned int *fanpcnt);
/*
* Query the current GPU speed from the CUDA device ID
*/
int wrap_nvml_get_clock(wrap_nvml_handle *nvmlh,
int gpuindex, int clktype, unsigned int *freq);
/*
* Query the current GPU power usage in millwatts from the CUDA device ID
*

Loading…
Cancel
Save