add gpu usage

gpu.go

@@ -2,7 +2,6 @@ package main
 
 import (
 	"fmt"
-	"log"
 
 	"github.com/NVIDIA/go-nvml/pkg/nvml"
 )
@@ -12,18 +11,18 @@ import (
 func getGPU() ([]DeviceInfo, error) {
 	ret := nvml.Init()
 	if ret != nvml.SUCCESS {
-		log.Fatalf("Unable to initialize NVML: %v", nvml.ErrorString(ret))
+		return nil, fmt.Errorf(ret)
 	}
 	defer func() {
 		ret := nvml.Shutdown()
 		if ret != nvml.SUCCESS {
-			log.Fatalf("Unable to shutdown NVML: %v", nvml.ErrorString(ret))
+			return nil, fmt.Errorf(err)
 		}
 	}()
 
 	count, ret := nvml.DeviceGetCount()
 	if ret != nvml.SUCCESS {
-		log.Fatalf("Unable to get device count: %v", nvml.ErrorString(ret))
+		return nil, fmt.Errorf(err)
 	}
 
 	res := make([]DeviceInfo, 0, count)
@@ -32,7 +31,7 @@ func getGPU() ([]DeviceInfo, error) {
 
 		device, ret := nvml.DeviceGetHandleByIndex(i)
 		if ret != nvml.SUCCESS {
-			log.Fatalf("Unable to get device at index %d: %v", i, nvml.ErrorString(ret))
+			return nil, fmt.Errorf(err)
 		}
 
 		// memory, err := device.GetMemoryInfo()
@@ -47,12 +46,13 @@ func getGPU() ([]DeviceInfo, error) {
 
 		name, err := device.GetName()
 		if err != nvml.SUCCESS {
-			panic(err)
+			return nil, fmt.Errorf(err)
 		}
 
 		memory, err := device.GetMemoryInfo()
 		if err != nvml.SUCCESS {
-			fmt.Println(err)
+			return nil, fmt.Errorf(err)
+			//fmt.Println(err)
 		}
 
 		//fmt.Printf(" %v\n", proc)
@@ -70,14 +70,16 @@ func getGPU() ([]DeviceInfo, error) {
 
 		if v, ok := comput[name]; ok {
 			c = v
-		} else {
-			fmt.Println("can not find the model '" + name + "'")
-
-			for k, v := range comput {
-				fmt.Println(k, v)
-			}
 		}
 
+		// } else {
+		// 	fmt.Println("can not find the model '" + name + "'")
+
+		// 	for k, v := range comput {
+		// 		fmt.Println(k, v)
+		// 	}
+		// }
+
 		e = DeviceInfo{
 			Type:  fmt.Sprintf("GPU-%d", i),
 			Model: name,
@@ -93,6 +95,74 @@ func getGPU() ([]DeviceInfo, error) {
 
 }
 
+func getGpuUsage() ([]DeviceInfo, error) {
+
+	ret := nvml.Init()
+	if ret != nvml.SUCCESS {
+		return nil, fmt.Errorf(ret)
+	}
+	defer func() {
+		ret := nvml.Shutdown()
+		if ret != nvml.SUCCESS {
+			return nil, fmt.Errorf(ret)
+			//log.Fatalf("Unable to shutdown NVML: %v", nvml.ErrorString(ret))
+		}
+	}()
+
+	count, ret := nvml.DeviceGetCount()
+	if ret != nvml.SUCCESS {
+		return nil, fmt.Errorf(ret)
+		//log.Fatalf("Unable to get device count: %v", nvml.ErrorString(ret))
+	}
+
+	res := make([]DeviceInfo, 0, count)
+
+	for i := 0; i < count; i++ {
+
+		device, ret := nvml.DeviceGetHandleByIndex(i)
+		if ret != nvml.SUCCESS {
+			return nil, fmt.Errorf(ret)
+			//log.Fatalf("Unable to get device at index %d: %v", i, nvml.ErrorString(ret))
+		}
+
+		name, ret := device.GetName()
+		if ret != nvml.SUCCESS {
+			return nil, fmt.Errorf(ret)
+			//panic(err)
+		}
+
+		utilization, ret := device.GetUtilizationRates()
+
+		if ret != nvml.SUCCESS {
+			return nil, fmt.Errorf(ret)
+		}
+
+		e := DeviceInfo{
+			Type:  fmt.Sprintf("GPU-%d", i),
+			Model: name,
+			Param: "gpu usage",
+			Power: utilization.Gpu,
+		}
+
+		res = append(res, e)
+
+		e = DeviceInfo{
+			Type:  fmt.Sprintf("GPU-%d", i),
+			Model: name,
+			Param: "gpu mem usage",
+			Power: utilization.Memory,
+		}
+
+		res = append(res, e)
+
+		//fmt.Println("utilization", utilization)
+
+	}
+
+	return res, nil
+
+}
+
 var comput map[string]uint64
 
 //NVIDIA GeForce RTX 3080,

main.go

@@ -52,6 +52,18 @@ func main() {
 
 	app.Get("/hw/used", func(c fiber.Ctx) error {
 
+		res := make([]DeviceInfo, 0)
+
+		gpu, err := getGpuUsage()
+
+		if err != nil {
+			c.SendString("getGpuUsage err: " + err.Error())
+		}
+
+		res = append(res, gpu...)
+
+		return c.JSON(res)
+
 		res := make([]DeviceInfo, 0)
 		return c.JSON(res)