|
|
@ -418,26 +418,6 @@ int nvml_reset_clocks(nvml_handle *nvmlh, int dev_id) |
|
|
|
return ret; |
|
|
|
return ret; |
|
|
|
} |
|
|
|
} |
|
|
|
|
|
|
|
|
|
|
|
int nvml_get_clocks(nvml_handle *nvmlh, int dev_id, unsigned int *core, unsigned int *mem) |
|
|
|
|
|
|
|
{ |
|
|
|
|
|
|
|
int ret = 0; |
|
|
|
|
|
|
|
nvmlReturn_t rc; |
|
|
|
|
|
|
|
uint32_t gpu_clk = 0, mem_clk = 0; |
|
|
|
|
|
|
|
int n = nvmlh->cuda_nvml_device_id[dev_id]; |
|
|
|
|
|
|
|
if (n < 0 || n >= nvmlh->nvml_gpucount) |
|
|
|
|
|
|
|
return -ENODEV; |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
if (nvmlh->nvmlDeviceGetClockInfo) { |
|
|
|
|
|
|
|
rc = nvmlh->nvmlDeviceGetClockInfo(nvmlh->devs[n], NVML_CLOCK_SM, &gpu_clk); |
|
|
|
|
|
|
|
rc = nvmlh->nvmlDeviceGetClockInfo(nvmlh->devs[n], NVML_CLOCK_MEM, &mem_clk); |
|
|
|
|
|
|
|
if (rc == NVML_SUCCESS) { |
|
|
|
|
|
|
|
*core = gpu_clk; *mem = mem_clk; |
|
|
|
|
|
|
|
return 1; |
|
|
|
|
|
|
|
} |
|
|
|
|
|
|
|
} |
|
|
|
|
|
|
|
return ret; |
|
|
|
|
|
|
|
} |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
/**
|
|
|
|
* Set power state of a device (9xx) |
|
|
|
* Set power state of a device (9xx) |
|
|
|
* Code is similar as clocks one, which allow the change of the pstate |
|
|
|
* Code is similar as clocks one, which allow the change of the pstate |
|
|
@ -663,7 +643,8 @@ int nvml_get_current_clocks(int cudaindex, uint32_t *graphics_clock, uint32_t *m |
|
|
|
{ |
|
|
|
{ |
|
|
|
nvmlReturn_t rc; |
|
|
|
nvmlReturn_t rc; |
|
|
|
int gpuindex = hnvml->cuda_nvml_device_id[cudaindex]; |
|
|
|
int gpuindex = hnvml->cuda_nvml_device_id[cudaindex]; |
|
|
|
if (gpuindex < 0 || gpuindex >= hnvml->nvml_gpucount) return -1; |
|
|
|
if (gpuindex < 0 || gpuindex >= hnvml->nvml_gpucount) return -ENODEV; |
|
|
|
|
|
|
|
if (!hnvml->nvmlDeviceGetClockInfo) return -ENOSYS; |
|
|
|
|
|
|
|
|
|
|
|
rc = hnvml->nvmlDeviceGetClockInfo(hnvml->devs[gpuindex], NVML_CLOCK_SM, graphics_clock); |
|
|
|
rc = hnvml->nvmlDeviceGetClockInfo(hnvml->devs[gpuindex], NVML_CLOCK_SM, graphics_clock); |
|
|
|
if (rc != NVML_SUCCESS) return -1; |
|
|
|
if (rc != NVML_SUCCESS) return -1; |
|
|
@ -2094,7 +2075,6 @@ extern int num_cpus; |
|
|
|
|
|
|
|
|
|
|
|
void *monitor_thread(void *userdata) |
|
|
|
void *monitor_thread(void *userdata) |
|
|
|
{ |
|
|
|
{ |
|
|
|
//struct thr_info *mythr = (struct thr_info *)userdata;
|
|
|
|
|
|
|
|
int thr_id = -1; |
|
|
|
int thr_id = -1; |
|
|
|
|
|
|
|
|
|
|
|
while (!abort_flag && !opt_quiet) |
|
|
|
while (!abort_flag && !opt_quiet) |
|
|
@ -2104,9 +2084,9 @@ void *monitor_thread(void *userdata) |
|
|
|
struct cgpu_info *cgpu = &thr_info[thr_id].gpu; |
|
|
|
struct cgpu_info *cgpu = &thr_info[thr_id].gpu; |
|
|
|
int dev_id = cgpu->gpu_id; cudaSetDevice(dev_id); |
|
|
|
int dev_id = cgpu->gpu_id; cudaSetDevice(dev_id); |
|
|
|
|
|
|
|
|
|
|
|
//applog(LOG_BLUE, "sampling device %d", dev_id);
|
|
|
|
|
|
|
|
if (hnvml != NULL && cgpu) |
|
|
|
if (hnvml != NULL && cgpu) |
|
|
|
{ |
|
|
|
{ |
|
|
|
|
|
|
|
char khw[32] = { 0 }; |
|
|
|
uint64_t clock = 0, mem_clock = 0; |
|
|
|
uint64_t clock = 0, mem_clock = 0; |
|
|
|
uint32_t fanpercent = 0, power = 0; |
|
|
|
uint32_t fanpercent = 0, power = 0; |
|
|
|
double tempC = 0, khs_per_watt = 0; |
|
|
|
double tempC = 0, khs_per_watt = 0; |
|
|
@ -2117,7 +2097,7 @@ void *monitor_thread(void *userdata) |
|
|
|
|
|
|
|
|
|
|
|
do { |
|
|
|
do { |
|
|
|
uint32_t tmp_clock, tmp_memclock; |
|
|
|
uint32_t tmp_clock, tmp_memclock; |
|
|
|
nvml_get_current_clocks(device_map[thr_id], &tmp_clock, &tmp_memclock); |
|
|
|
nvml_get_current_clocks(dev_id, &tmp_clock, &tmp_memclock); |
|
|
|
clock += tmp_clock; |
|
|
|
clock += tmp_clock; |
|
|
|
mem_clock += tmp_memclock; |
|
|
|
mem_clock += tmp_memclock; |
|
|
|
tempC += gpu_temp(cgpu); |
|
|
|
tempC += gpu_temp(cgpu); |
|
|
@ -2137,15 +2117,16 @@ void *monitor_thread(void *userdata) |
|
|
|
cgpu->monitor.gpu_memclock = (uint32_t) (mem_clock/counter); |
|
|
|
cgpu->monitor.gpu_memclock = (uint32_t) (mem_clock/counter); |
|
|
|
|
|
|
|
|
|
|
|
if (power) { |
|
|
|
if (power) { |
|
|
|
// todo: handle units
|
|
|
|
khs_per_watt = stats_get_speed(thr_id, thr_hashrates[thr_id]); |
|
|
|
khs_per_watt = stats_get_speed(thr_id, thr_hashrates[thr_id]) / ((double)power / counter); |
|
|
|
khs_per_watt = khs_per_watt / ((double)power / counter); |
|
|
|
|
|
|
|
format_hashrate(khs_per_watt * 1000, khw); |
|
|
|
|
|
|
|
if (strlen(khw)) khw[strlen(khw)-1] = 'W'; |
|
|
|
} |
|
|
|
} |
|
|
|
|
|
|
|
|
|
|
|
// todo: not shown on decred
|
|
|
|
|
|
|
|
if (opt_hwmonitor && (time(NULL) - cgpu->monitor.tm_displayed) > 60) { |
|
|
|
if (opt_hwmonitor && (time(NULL) - cgpu->monitor.tm_displayed) > 60) { |
|
|
|
gpulog(LOG_INFO, thr_id, "%u MHz %.3f kH/W %uW %uC FAN %u%%", |
|
|
|
gpulog(LOG_INFO, thr_id, "%u MHz %s %uW %uC FAN %u%%", |
|
|
|
cgpu->monitor.gpu_clock/*, cgpu->monitor.gpu_memclock*/, |
|
|
|
cgpu->monitor.gpu_clock/*, cgpu->monitor.gpu_memclock*/, |
|
|
|
khs_per_watt, cgpu->monitor.gpu_power / 1000, |
|
|
|
khw, cgpu->monitor.gpu_power / 1000, |
|
|
|
cgpu->monitor.gpu_temp, cgpu->monitor.gpu_fan |
|
|
|
cgpu->monitor.gpu_temp, cgpu->monitor.gpu_fan |
|
|
|
); |
|
|
|
); |
|
|
|
cgpu->monitor.tm_displayed = (uint32_t)time(NULL); |
|
|
|
cgpu->monitor.tm_displayed = (uint32_t)time(NULL); |
|
|
|