你好，我这边在AWS F1机器上尝试这个服务。整个log如下。 Note : kubectl version is v.16.3，so I had modified the yaml file. 使用kubectl create, 用一般的yaml，或者aws folder下面的yaml，两个服务都能起来没有报错，但是打log看貌似都不像正常可以工作的样子。请帮忙确认是否这个状态OK?

ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl get pod -n kube-system NAME READY STATUS RESTARTS AGE coredns-67c766df46-k8kk7 1/1 Running 1 46h coredns-67c766df46-kdlw5 1/1 Running 1 46h etcd-minikube 1/1 Running 1 46h kube-addon-manager-minikube 1/1 Running 1 46h kube-apiserver-minikube 1/1 Running 1 46h kube-controller-manager-minikube 1/1 Running 1 46h kube-proxy-xlng4 1/1 Running 1 47h kube-scheduler-minikube 1/1 Running 1 46h storage-provisioner 1/1 Running 2 46h ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl create -f fpga-device-plugin.ymldaemonset.apps/fpga-device-plugin-daemonset created ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl get pod -n kube-system NAME READY STATUS RESTARTS AGE coredns-67c766df46-k8kk7 1/1 Running 1 47h coredns-67c766df46-kdlw5 1/1 Running 1 47h etcd-minikube 1/1 Running 1 46h fpga-device-plugin-daemonset-nvzgp 1/1 Running 0 22s kube-addon-manager-minikube 1/1 Running 1 46h kube-apiserver-minikube 1/1 Running 1 46h kube-controller-manager-minikube 1/1 Running 1 46h kube-proxy-xlng4 1/1 Running 1 47h kube-scheduler-minikube 1/1 Running 1 46h storage-provisioner 1/1 Running 2 47h ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl logs fpga-device-plugin-daemonset-nvzgp -n kube-system time="2019-12-15T06:25:14Z" level=info msg="Starting FS watcher." time="2019-12-15T06:25:14Z" level=info msg="Starting OS watcher." ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ cd aws ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl create -f aws-fpga-device-plugin.yaml Error from server (AlreadyExists): error when creating "aws-fpga-device-plugin.yaml": daemonsets.apps "fpga-device-plugin-daemonset" already exists ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ vi aws-fpga-device-plugin.yaml ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl create -f aws-fpga-device-plugin.yaml daemonset.apps/aws-fpga-device-plugin-daemonset created ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl get pod -n kube-system NAME READY STATUS RESTARTS AGE aws-fpga-device-plugin-daemonset-d9tgj 1/1 Running 0 20s coredns-67c766df46-k8kk7 1/1 Running 1 47h coredns-67c766df46-kdlw5 1/1 Running 1 47h etcd-minikube 1/1 Running 1 47h fpga-device-plugin-daemonset-nvzgp 1/1 Running 0 3m21s kube-addon-manager-minikube 1/1 Running 1 47h kube-apiserver-minikube 1/1 Running 1 47h kube-controller-manager-minikube 1/1 Running 1 47h kube-proxy-xlng4 1/1 Running 1 47h kube-scheduler-minikube 1/1 Running 1 47h storage-provisioner 1/1 Running 2 47h ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl logs aws-fpga-device-plugin-daemonset-d9tgj -n kube-system time="2019-12-15T06:28:09Z" level=info msg="Starting FS watcher." time="2019-12-15T06:28:09Z" level=info msg="Starting OS watcher." ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$

Aws 目录下面的那个可能没有编好的docker 镜像，我记不起来了。所以你也可能需要自己编。主目录下面的需要xrt 版本2019.2，aws官方还没有，需要自己编，过程比编一般xrt要复杂一些，不知道什么地方有说明，如果你想自己编，我回头给你一个步骤说明

On Sat, Dec 14, 2019, 22:46 Liang Ma notifications@github.com wrote:

你好，我这边在AWS F1机器上尝试这个服务。整个log如下。 Note : kubectl version is v.16.3，so I had modified the yaml file. 使用kubectl create, 用一般的yaml，或者aws folder下面的yaml，两个服务都能起来没有报错，但是打log看貌似都不像正常可以工作的样子。请帮忙确认是否这个状态OK?

ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl get pod -n kube-system NAME READY STATUS RESTARTS AGE coredns-67c766df46-k8kk7 1/1 Running 1 46h coredns-67c766df46-kdlw5 1/1 Running 1 46h etcd-minikube 1/1 Running 1 46h kube-addon-manager-minikube 1/1 Running 1 46h kube-apiserver-minikube 1/1 Running 1 46h kube-controller-manager-minikube 1/1 Running 1 46h kube-proxy-xlng4 1/1 Running 1 47h kube-scheduler-minikube 1/1 Running 1 46h storage-provisioner 1/1 Running 2 46h ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl create -f fpga-device-plugin.ymldaemonset.apps/fpga-device-plugin-daemonset created ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl get pod -n kube-system NAME READY STATUS RESTARTS AGE coredns-67c766df46-k8kk7 1/1 Running 1 47h coredns-67c766df46-kdlw5 1/1 Running 1 47h etcd-minikube 1/1 Running 1 46h fpga-device-plugin-daemonset-nvzgp 1/1 Running 0 22s kube-addon-manager-minikube 1/1 Running 1 46h kube-apiserver-minikube 1/1 Running 1 46h kube-controller-manager-minikube 1/1 Running 1 46h kube-proxy-xlng4 1/1 Running 1 47h kube-scheduler-minikube 1/1 Running 1 46h storage-provisioner 1/1 Running 2 47h ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl logs fpga-device-plugin-daemonset-nvzgp -n kube-system time="2019-12-15T06:25:14Z" level=info msg="Starting FS watcher." time="2019-12-15T06:25:14Z" level=info msg="Starting OS watcher." ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ cd aws ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl create -f aws-fpga-device-plugin.yaml Error from server (AlreadyExists): error when creating "aws-fpga-device-plugin.yaml": daemonsets.apps "fpga-device-plugin-daemonset" already exists ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ vi aws-fpga-device-plugin.yaml ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl create -f aws-fpga-device-plugin.yaml daemonset.apps/aws-fpga-device-plugin-daemonset created ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl get pod -n kube-system NAME READY STATUS RESTARTS AGE aws-fpga-device-plugin-daemonset-d9tgj 1/1 Running 0 20s coredns-67c766df46-k8kk7 1/1 Running 1 47h coredns-67c766df46-kdlw5 1/1 Running 1 47h etcd-minikube 1/1 Running 1 47h fpga-device-plugin-daemonset-nvzgp 1/1 Running 0 3m21s kube-addon-manager-minikube 1/1 Running 1 47h kube-apiserver-minikube 1/1 Running 1 47h kube-controller-manager-minikube 1/1 Running 1 47h kube-proxy-xlng4 1/1 Running 1 47h kube-scheduler-minikube 1/1 Running 1 47h storage-provisioner 1/1 Running 2 47h ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl logs aws-fpga-device-plugin-daemonset-d9tgj -n kube-system time="2019-12-15T06:28:09Z" level=info msg="Starting FS watcher." time="2019-12-15T06:28:09Z" level=info msg="Starting OS watcher." ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM2KV4HXRIRRYBZR7W3QYXHDTA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IARNBCQ, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM5EBIPQJSA4CYWEALTQYXHDTANCNFSM4J26WL3A .

AWS目录下面的好像也有docker 镜像。你的log显示的好像daemonset还是主目录下创建的，请查看一下docker 镜像是tag20190627。另外k8s v16需要像主目录下的一样编辑一下那个yaml文件

On Sun, Dec 15, 2019, 06:01 huazhuo xu huazhuo.xu@gmail.com wrote:

Aws 目录下面的那个可能没有编好的docker 镜像，我记不起来了。所以你也可能需要自己编。主目录下面的需要xrt 版本2019.2，aws官方还没有，需要自己编，过程比编一般xrt要复杂一些，不知道什么地方有说明，如果你想自己编，我回头给你一个步骤说明

On Sat, Dec 14, 2019, 22:46 Liang Ma notifications@github.com wrote:

你好，我这边在AWS F1机器上尝试这个服务。整个log如下。 Note : kubectl version is v.16.3，so I had modified the yaml file. 使用kubectl create, 用一般的yaml，或者aws folder下面的yaml，两个服务都能起来没有报错，但是打log看貌似都不像正常可以工作的样子。请帮忙确认是否这个状态OK?

ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl get pod -n kube-system NAME READY STATUS RESTARTS AGE coredns-67c766df46-k8kk7 1/1 Running 1 46h coredns-67c766df46-kdlw5 1/1 Running 1 46h etcd-minikube 1/1 Running 1 46h kube-addon-manager-minikube 1/1 Running 1 46h kube-apiserver-minikube 1/1 Running 1 46h kube-controller-manager-minikube 1/1 Running 1 46h kube-proxy-xlng4 1/1 Running 1 47h kube-scheduler-minikube 1/1 Running 1 46h storage-provisioner 1/1 Running 2 46h ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl create -f fpga-device-plugin.ymldaemonset.apps/fpga-device-plugin-daemonset created ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl get pod -n kube-system NAME READY STATUS RESTARTS AGE coredns-67c766df46-k8kk7 1/1 Running 1 47h coredns-67c766df46-kdlw5 1/1 Running 1 47h etcd-minikube 1/1 Running 1 46h fpga-device-plugin-daemonset-nvzgp 1/1 Running 0 22s kube-addon-manager-minikube 1/1 Running 1 46h kube-apiserver-minikube 1/1 Running 1 46h kube-controller-manager-minikube 1/1 Running 1 46h kube-proxy-xlng4 1/1 Running 1 47h kube-scheduler-minikube 1/1 Running 1 46h storage-provisioner 1/1 Running 2 47h ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl logs fpga-device-plugin-daemonset-nvzgp -n kube-system time="2019-12-15T06:25:14Z" level=info msg="Starting FS watcher." time="2019-12-15T06:25:14Z" level=info msg="Starting OS watcher." ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ cd aws ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl create -f aws-fpga-device-plugin.yaml Error from server (AlreadyExists): error when creating "aws-fpga-device-plugin.yaml": daemonsets.apps "fpga-device-plugin-daemonset" already exists ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ vi aws-fpga-device-plugin.yaml ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl create -f aws-fpga-device-plugin.yaml daemonset.apps/aws-fpga-device-plugin-daemonset created ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl get pod -n kube-system NAME READY STATUS RESTARTS AGE aws-fpga-device-plugin-daemonset-d9tgj 1/1 Running 0 20s coredns-67c766df46-k8kk7 1/1 Running 1 47h coredns-67c766df46-kdlw5 1/1 Running 1 47h etcd-minikube 1/1 Running 1 47h fpga-device-plugin-daemonset-nvzgp 1/1 Running 0 3m21s kube-addon-manager-minikube 1/1 Running 1 47h kube-apiserver-minikube 1/1 Running 1 47h kube-controller-manager-minikube 1/1 Running 1 47h kube-proxy-xlng4 1/1 Running 1 47h kube-scheduler-minikube 1/1 Running 1 47h storage-provisioner 1/1 Running 2 47h ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl logs aws-fpga-device-plugin-daemonset-d9tgj -n kube-system time="2019-12-15T06:28:09Z" level=info msg="Starting FS watcher." time="2019-12-15T06:28:09Z" level=info msg="Starting OS watcher." ubuntu@ip-:~ /faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM2KV4HXRIRRYBZR7W3QYXHDTA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IARNBCQ, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM5EBIPQJSA4CYWEALTQYXHDTANCNFSM4J26WL3A .

xilinx主目录下的yml: fpga-device-plugin.yml aws目录下的ymal: aws-fpga-device-plugin.yaml xilinx镜像: xilinxatg/xilinx_k8s_fpga_plugin:latest aws镜像: xilinxatg/aws_k8s_fpga_plugin:06272019

我也diff了一下两个yml的不同，见下图：

请教一下还有什么地方需要注意？

无论哪个服务，打log出来，只有这两行： time="2019-12-15T06:28:09Z" level=info msg="Starting FS watcher." time="2019-12-15T06:28:09Z" level=info msg="Starting OS watcher."

$kubectl get node 看不到fpga resource. NAME STATUS ROLES AGE VERSION minikube Ready master 4d1h v1.16.2

首先，确保有fpga节点， root@ip-172-31-70-242:~# lspci |grep f010 00:1d.0 Memory controller: Amazon.com, Inc. Device f010

第二，尝试手工先把插件的镜像抓下来。

docker pull xilinxatg/aws_k8s_fpga_plugin:06272019

第三，如果曾用主目录下的那个创建过daemonset，先把它删掉

kubectl get ds -n kube-system

kubectl delete ds xxx -n kube-system

第四，用aws目录下面的yaml创建。v16的话，那个yaml文件需要参照主目录下的修改一下。你的yaml文件没问题。

第五，get node 看不见fpga信息，需要describe node

On Tue, Dec 17, 2019 at 1:07 AM Liang Ma notifications@github.com wrote:

无论哪个服务，打log出来，只有这两行： time="2019-12-15T06:28:09Z" level=info msg="Starting FS watcher." time="2019-12-15T06:28:09Z" level=info msg="Starting OS watcher."

$kubectl get node 看不到fpga resource. NAME STATUS ROLES AGE VERSION minikube Ready master 4d1h v1.16.2

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM25TP2V6OHFYNQLQTDQZCJDPA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEHBVRVA#issuecomment-566450388, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM64PLJKOT5ZXPWTHZTQZCJDPANCNFSM4J26WL3A .

您好，非常感谢！

lspci | grep f010 没有设备，lspci 可以看到两个相关设备： 00:1d.0 Memory controller: Amazon.com, Inc. Device 1042 00:1e.0 Memory controller: Amazon.com, Inc. Device 1041
按照推荐流程，daemonset服务起来以后，打log看到如下报错： time="2019-12-18T03:03:12Z" level=info msg="Starting FS watcher." time="2019-12-18T03:03:12Z" level=info msg="Starting OS watcher." time="2019-12-18T03:03:12Z" level=error msg="Error to get FPGA devices: Can't read folder /sys/bus/pci/devices/0000:00:1d.0/drm \n"

cd /sys/bus/pci/devices/0000:00:1d.0/, 没有drm 目录。

我看到fpga_aws.go 里面有定义一些const在这里。是否这里需要做一些更改来适配我这个AWS F1?

另，我用aws fpga cli （比如 fpga-load-local-image）可以load aws 的AFI，运行无问题。 ———————————————————————————————— 额外一个问题： import ( 。。。 pluginapi "k8s.io/kubernetes/pkg/kubelet/apis/deviceplugin/v1beta1"

Yaml里面plugin版本apiVersion:apps/v1，是否代码里也需要更改？

1042和f010是同一个设备，没load afi是1042， load完就是f010 你登陆到插件的pod里面看看，sysfs是不是mount了，

mount |grep sysfs

sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime) 容器里面默认sysfs是mount成只读的而且应该可以看见，

ls /sys/bus/pci/devices/0000\:00\:1d.0/drm/

card0 renderD128

On Tue, Dec 17, 2019 at 7:13 PM Liang Ma notifications@github.com wrote:

您好，非常感谢！

1.

lspci | grep f010 没有设备，lspci 可以看到两个相关设备： 00:1d.0 Memory controller: Amazon.com, Inc. Device 1042 00:1e.0 Memory controller: Amazon.com, Inc. Device 1041 2.

按照推荐流程，daemonset服务起来以后，打log看到如下报错： time="2019-12-18T03:03:12Z" level=info msg="Starting FS watcher." time="2019-12-18T03:03:12Z" level=info msg="Starting OS watcher." time="2019-12-18T03:03:12Z" level=error msg="Error to get FPGA devices: Can't read folder /sys/bus/pci/devices/0000:00:1d.0/drm \n"

Q1: 这个会否有权限问题？ Q2: 根据lspci的结果，你的命令里面Device f010，我这边aws上面是 Device 1042 和 Device 1041。请问是否代码某处又hardcode 要用f010?

另，我用aws fpga cli可以log aws 的AFI。

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM7YHSDOR5W3GPP6MD3QZGIM7A5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEHEWUBQ#issuecomment-566848006, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM6PYHYSXRM443TDAQTQZGIM7ANCNFSM4J26WL3A .

OK，多谢。

我试了一下，我这边没有load afi的时候是1042， fpga-load-local-image 成功以后，是f000. 这个重要吗？我曾经尝试了把aws-fpga.go里面的f010改成f000，没有什么作用。
“登陆到插件的pod里面看看”：mount 是OK的，但是/drm确实没有： $~/faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl exec aws-fpga-device-plugin-daemonset-wp4sv -i -t -n kube-system bash root@aws-fpga-device-plugin-daemonset-wp4sv:/# mount | grep sysfs sysfs on /sys type sysfs (ro,nosuid,nodev,noexec,relatime) root@aws-fpga-device-plugin-daemonset-wp4sv:/# ls /sys/bus/pci/devices/0000\:00\:1d.0/drm/ ls: cannot access '/sys/bus/pci/devices/0000:00:1d.0/drm/': No such file or directory

我注意到我的plugin的NODE SELECTOR= none ：是否这个有影响？ $ kubectl get ds -n kube-system NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE aws-fpga-device-plugin-daemonset 1 1 1 1 1 101m kube-proxy 1 1 1 1 1 beta.kubernetes.io/os=linux 6d2h

我没见过f000。不过目前你的碰到的问题跟这个没关系。你看看在vm里面有drm吗？container里面的sysfs是host map过去的。 aws#lsmod |grep drm aws# ls /sys/bus/pci/devices/0000:00:1d.0/drm/

On Thu, Dec 19, 2019 at 1:51 AM Liang Ma notifications@github.com wrote:

OK，多谢。我试了一下，我这边没有load afi的时候是1042， fpga-load-local-image 成功以后，是f000. 这个重要吗？

“登陆到插件的pod里面看看”：mount 是OK的，但是/drm确实没有：

$~/faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk/aws$ kubectl exec aws-fpga-device-plugin-daemonset-wp4sv -i -t -n kube-system bash root@aws-fpga-device-plugin-daemonset-wp4sv:/# mount | grep sysfs sysfs on /sys type sysfs (ro,nosuid,nodev,noexec,relatime) root@aws-fpga-device-plugin-daemonset-wp4sv:/# ls /sys/bus/pci/devices/0000:00:1d.0/drm/ ls: cannot access '/sys/bus/pci/devices/0000:00:1d.0/drm/': No such file or directory

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBMYOBVAJIECTXGP2QH3QZM7YPA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEHJBWWA#issuecomment-567417688, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM6SUTICYZ36Y4WQFPTQZM7YPANCNFSM4J26WL3A .

没有的，ls /sys/bus/pci/devices/0000:00:1d.0/ 存在有不少内容。但是没有drm目录： ~$ ls /sys/bus/pci/devices/0000:00:1d.0/ broken_parity_status config current_link_speed d3cold_allowed dma_mask_bits enable irq local_cpus max_link_width msi_bus power rescan resource0 resource2 resource4 revision subsystem_device uevent class consistent_dma_mask_bits current_link_width device driver_override firmware_node local_cpulist max_link_speed modalias numa_node remove resource resource1 resource2_wc resource4_wc subsystem subsystem_vendor vendor 请教这个drm的用途？

Xrt似乎没有安装。看看f1上xocl加载没有，/opt/xilinx有没有内容

On Thu, Dec 19, 2019, 17:43 Liang Ma notifications@github.com wrote:

没有的，ls /sys/bus/pci/devices/0000:00:1d.0/ 存在有不少内容。但是没有drm目录： ~$ ls /sys/bus/pci/devices/0000:00:1d.0/ broken_parity_status config current_link_speed d3cold_allowed dma_mask_bits enable irq local_cpus max_link_width msi_bus power rescan resource0 resource2 resource4 revision subsystem_device uevent class consistent_dma_mask_bits current_link_width device driver_override firmware_node local_cpulist max_link_speed modalias numa_node remove resource resource1 resource2_wc resource4_wc subsystem subsystem_vendor vendor

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBMYIIBJV54EBLVAJE53QZQPLRA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEHLTHDA#issuecomment-567751564, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM3GN6SRYCQECF52ETTQZQPLRANCNFSM4J26WL3A .

看看f1上xocl加载没有：请教这个如何操作？
/opt/xilinx ：没有。/opt 下面只有一个目录：containerd

对了，这台F1是没有安装Vivado的。source hdk_setup 报错。但是source sdk_setup是pass的，我本来以为这样xrt就安装OK了。

Done with SDK install. INFO: sdk_setup.sh PASSED ERROR: VIVADO_TOOL_VERSION ENV variable is not set. ERROR: ENV Variable VIVADO_TOOL_VERSION needs to be set for runtime usage. ERROR: If AFI was generated using V2019.1 tools use the command : export VIVADO_TOOL_VERSION=2019.1 ERROR: If AFI was generated using V2018.3 tools use the command : export VIVADO_TOOL_VERSION=2018.3 ERROR: If AFI was generated using V2018.2 tools use the command : export VIVADO_TOOL_VERSION=2018.2 ERROR: If AFI was generated using V2017.4 tools use the command : export VIVADO_TOOL_VERSION=2017.4 ERROR: Please set VIVADO_TOOL_VERSION to the correct value and re-run script.

如何确认安装Xilix xrt (remotely, by command line, install to AWS F1)，有没有guidance 可以提供？是否可以follow 这里的安装：https://github.com/Xilinx/XRT

问题：XRT如果没有安装，为什么aws AFI可以load并运行呢?或者AWS AFI是自己customized RT?

您好，多谢建议。对于fpga和k8s我都不是很熟悉，抱歉可能一步一步和您check.

折腾了小半天，我应该是把XRT安装起来了。问题1：XRT没有安装，为什么aws AFI可以load并运行呢? 是否AWS F1机器上面是自己customized Runtime?)

现在，kubectl logs可以看到如下消息： $kubectl logs aws-fpga-device-plugin-daemonset-n7j8f -n kube-system time="2019-12-20T10:08:44Z" level=info msg="Starting FS watcher." time="2019-12-20T10:08:44Z" level=info msg="Starting OS watcher." time="2019-12-20T10:08:44Z" level=info msg="Starting to serve on /var/lib/kubelet/device-plugins/xilinx_aws-vu9p-f1-04261818_dynamic_5_0-0-fpga.sock" 2019/12/20 10:08:44 grpc: Server.Serve failed to create ServerTransport: connection error: desc = "transport: write unix /var/lib/kubelet/device-plugins/xilinx_aws-vu9p-f1-04261818_dynamic_5_0-0-fpga.sock->@: write: broken pipe" time="2019-12-20T10:08:44Z" level=info msg="Registered device plugin with Kubelet xilinx.com/fpga-xilinx_aws-vu9p-f1-04261818_dynamic_5_0-0" time="2019-12-20T10:08:44Z" level=info msg="Sending 1 device(s) [&Device{ID:0000:00:1d.0,Health:Healthy,}] to kubelet"

问题2：这之后没有其它log，正常状况应该是有“msg="Receiving request 1"的消息吗？

现在进一步查看node，只有一个master node, 没有fpga node，这个正常吗？问题3：没有fpga node，如上create的fpga pod（aws-fpga-device-plugin-daemonset-n7j8f ），运行在哪里呢？

$kubectl get node -n kube-system NAME STATUS ROLES AGE VERSION minikube Ready master 7d2h v1.16.2

1 fpga 有两个pci function， mgmt PF 和user PF，aws 的fgpa*命令在操作mgmt， xrt安装完后加载xocl 驱动操作user。注意，xrt本身包含两个function的驱动，aws对fgpa有定制修改，不用xrt里面的xclmgmt。修改过mgmt主要负责管理，比如加载用户编写的bitstream。用户本身只关心user PF，插件也只把user PF分配给container

2 正常。插件应该运行了，部署一个用户pod就可以看见 "Receiving request 1" 3 你这是只包含一个node的cluster，所有的pod都运行在这个node上。 k8s中node相当于一个server，在这里就是vm，pod相当于container，plugin本身也运行在pod中，在每一个节点上都运行这样一个pod，报告那个节点上的fpga情况，并接受分配请求。

你编的xrt是上面版本？2019.2+ 的话就不要用aws目录下的插件了。2019.2之前的xrt只能用aws下的插件，而且还有缺陷，就是，用户pod必须部署成特权模式，才能下载afi。2019.2以后的xrt没有这个限制

On Fri, Dec 20, 2019 at 2:19 AM Liang Ma notifications@github.com wrote:

您好，感谢建议。折腾了半天，我应该是把XRT安装起来了。（问题1：XRT没有安装，为什么aws AFI可以load并运行呢? 是否AWS F1机器上面是自己customized Runtime?)

现在kubectl logs可以看到如下消息： kubectl logs aws-fpga-device-plugin-daemonset-n7j8f -n kube-system time="2019-12-20T10:08:44Z" level=info msg="Starting FS watcher." time="2019-12-20T10:08:44Z" level=info msg="Starting OS watcher." time="2019-12-20T10:08:44Z" level=info msg="Starting to serve on /var/lib/kubelet/device-plugins/xilinx_aws-vu9p-f1-04261818_dynamic_5_0-0-fpga.sock" 2019/12/20 10:08:44 grpc: Server.Serve failed to create ServerTransport: connection error: desc = "transport: write unix /var/lib/kubelet/device-plugins/xilinx_aws-vu9p-f1-04261818_dynamic_5_0-0-fpga.sock->@: write: broken pipe" time="2019-12-20T10:08:44Z" level=info msg="Registered device plugin with Kubelet xilinx.com/fpga-xilinx_aws-vu9p-f1-04261818_dynamic_5_0-0" time="2019-12-20T10:08:44Z" level=info msg="Sending 1 device(s) [&Device{ID:0000:00:1d.0,Health:Healthy,}] to kubelet"

问题2：这之后没有其它log，正常状况应该是有“msg="Receiving request 1"的消息吗？

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBMZHLUOW3SD24CTVOC3QZSL3RA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEHMQ5RA#issuecomment-567873220, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM7CAEVM5QUVED2JYNTQZSL3RANCNFSM4J26WL3A .

您好，多谢指教，很多信息啊，只是还要再消化一下。关于最后一点我有点疑惑，是否是说如果我用的XRT > 2019.2，我就不用aws的插件，而是用主目录下的插件？ anyway，我现在分别跑了主目录下的插件和aws下的插件：

主目录下插件，修改了部署之后mypod 是pending: kubectl get pods --all-namespaces NAMESPACE NAME READY STATUS RESTARTS AGE default mypod 0/1 Pending 0 2m34s kube-system fpga-device-plugin-daemonset-k7k2s 1/1 Running 0 2m43s

之后修改了mypod里面的resource limit, 重新部署就OK了。

但是这之后登陆到pod里面尝试运行helloworld，遇到如下状况： ubuntu@ip-172-31-22-109:~/faas/FPGA_as_a_Service/k8s-fpga-device-plugin/trunk$ kubectl exec -it lma-fpga-pod /bin/bash root@lma-fpga-pod:/# source /opt/xilinx/xrt/setup.sh XILINX_XRT : /opt/xilinx/xrt PATH : /opt/xilinx/xrt/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin LD_LIBRARY_PATH : /opt/xilinx/xrt/lib: root@lma-fpga-pod:/# xbutil scan ERROR: No card found //这里貌似fpga card资源没有没map到container里面. root@lma-fpga-pod:/tmp/alveo-u200/xilinx_u200_xdma_201830_1/test# ls bandwidth.xclbin kernel_bw.exe validate.exe verify.xclbin root@lma-fpga-pod:/tmp/alveo-u200/xilinx_u200_xdma_201830_1/test# ./validate.exe ./verify.xclbin bash: ./validate.exe: Permission denied //没有permission??

退出pod,在container外部运行xbutil scan可以看到fpga card，但是有如下警告： INFO: Found total 1 card(s), 0 are usable ........ [0] 0000:00:1d.0 xilinx_aws-vu9p-f1_dynamic_5_0(ts=0xabcd) user(inst=128) WARNING: card(s) marked by '' are not ready, is MPD runing? run 'systemctl status mpd' to check MPD details. ————————————————————————————————

aws目录下插件，create以后my-pod是ErrImagePull kubectl get pods --all-namespaces NAMESPACE NAME READY STATUS RESTARTS AGE default my-pod 0/1 ErrImagePull 0 12s kube-system aws-fpga-device-plugin-daemonset-fczwj 1/1 Running 0 21s describe mypod有如下error: Type Reason Age From Message

Normal Scheduled default-scheduler Successfully assigned default/my-pod to minikube Normal Pulling 39s (x3 over 94s) kubelet, minikube Pulling image "centos:bx" Warning Failed 32s (x3 over 86s) kubelet, minikube Failed to pull image "centos:bx": rpc error: code = Unknown desc = Error response from daemon: manifest for centos:bx not found: manifest unknown: manifest unknown

查看了mypod.ymal里面，image name 是centos:bx #user needs to build its own docker image 然后尝试自己。/build. 但是跳出来一堆和go相关的error. 类似如下： main.go:23:2: cannot find package "github.com/fsnotify/fsnotify" in any of: /usr/lib/go-1.13/src/github.com/fsnotify/fsnotify (from $GOROOT) /home/ubuntu/go/src/github.com/fsnotify/fsnotify (from $GOPATH) server.go:27:2: cannot find package "golang.org/x/net/context" in any of: /usr/lib/go-1.13/src/golang.org/x/net/context (from $GOROOT) /home/ubuntu/go/src/golang.org/x/net/context (from $GOPATH) ....... 其实这几个go 源代码和主目录下的是一样的吧？我在主目录下./build没有error..很纳闷。也请指教。

您好，新年好。抱歉问题还在没太多进展。额外有个问题3)，XRT在build完成以后有多个installation package: xrt_202010.2.5.0_18.04-aws.deb xrt_202010.2.5.0_18.04-xbtest.deb xrt_202010.2.5.0_18.04-azure.deb xrt_202010.2.5.0_18.04-xrt.deb xrt_202010.2.5.0_18.04-container.deb

请问一下怎么选取安装？貌似aws.deb是给aws 的pckage (但是我如果选择这个安装会报错）。 -xbtest.deb, -xrt.deb, -container.deb 各自有什么区别能否解释一下？实际我用 -xrt.deb, 安装可以成功，其他后续问题如前一个post里面#1， #2的描述。

在aws上面需要先安装xrt，再安装aws。另外，不知道你编的xrt是按照上面步骤完成的？编aws的xrt不是很直接，一般由aws负责，大致过程如下。

Git clone https://github.com/aws/aws-fpga.git
Cd aws-fpga/SDAccel/Runtime
Git clone https://github.com/xilinx/xrt.git
Mv xrt XRT_2019.1
XRT_2019.1/src/runtime_src/tools/scripts/xrtdeps.sh (install dependency pkgs)
scl enable devtoolset-6 bash （for centos only)
cd ../../.. (goto the top level of the aws-fpga workspace)
vi sdaccel_setup.sh (2019 is not supported by default)

85 RELEASE_VER=${RELEASE_VER:0:6}

86 RELEASE_VER="2019.1" <<<<<<< add this line

87 export RELEASE_VER=$RELEASE_VER

175 if [[ $RELEASE_VER =~ .2017.4. || $RELEASE_VER =~ .2018.2. || $RELEASE_VER =~ .2018.3. || $RELEASE_VER =~ .2019.1. ]]; then <<<< change this line to add 2019.1

source sdaccel_setup.sh
cd aws-fpga/SDAccel/Runtime/XRT_2019.1/build
./build.sh

安装完xrt和aws后，应该有一个mpd daemon在运行。

systemctl status mpd

xbutil scan

不会再有告警。

然后部署主目录下面的插件就可以了。在aws上面没有现成的验证xclbin可用。你需要自己编一个（得到一个后缀为awsxclbin的文件，这个文件很小，其实就是一个id，真正的xclbin文件存在S3上面，用户不可见。加载时通过id找到真正的文件）

前面的问题#1，告警是因为aws包没有安装，所以mpd没有运行。注意，xrt包在哪里是一样的，怎么编都可以。aws包要通过上面的步骤来，否则里面的aws插件找不到aws提供的一个库。不通过上面步骤也会生成一个aws包，只是用来内部模拟测试使用的。 fpga-verify那个container image不能在aws上面使用，因为里面没有针对aws上面fpga shell版本的xclbin文件。

问题#2，示例用到的mypod镜像没有在docker-hub上，所以返回pull request err。如前所述，你需要自己先弄一个可以在F1上运行的xclbin和应用，然后把它放在docker image里面，就可以在pod里面运行了。

On Thu, Jan 2, 2020 at 1:39 AM Liang Ma notifications@github.com wrote:

您好，新年好。抱歉问题还在没太多进展。额外有个问题3)，XRT在build完成以后有多个installation package: xrt_202010.2.5.0_18.04-aws.deb xrt_202010.2.5.0_18.04-xbtest.deb xrt_202010.2.5.0_18.04-azure.deb xrt_202010.2.5.0_18.04-xrt.deb xrt_202010.2.5.0_18.04-container.deb

请问一下怎么选取安装？貌似aws.deb是给aws 的pckage (但是我如果选择这个安装会报错）。 -xbtest.deb, -xrt.deb, -container.deb 各自有什么区别能否解释一下？实际我用 -xrt.deb, 安装可以成功，其他后续问题如前一个post里面#1 https://github.com/Xilinx/FPGA_as_a_Service/pull/1， #2的描述。

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM3DKVQ6OMBIQWHNDU3Q3WY4LA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEH56R6I#issuecomment-570157305, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM7LMHGDP5URSPPXT23Q3WY4LANCNFSM4J26WL3A .

您好：我先不去弄container这些了，先搞AWS和XRT: Follow up 您的建议（AWS support forum 上也给了类似建议，区别只是没有mv XRT XRT_2019.1这一步而已。我的问题好像仍旧..

vi sdaccel_setup.sh 的修改，这一步对我的没有帮助。因为source sdaccel_setup.sh 内部调用执行完sdk_setup.sh PASS 以后报告ERROR: XILINX_SDX variable not set and sdx not in the path--根本没有执行到修改添加的部分就结束退出。这里貌似是因为我没有安装SDAccel开发工具--但是理论上说对于部署来说这个应该不必要对吧？
后续执行安装完还是老问题，如下：安装过程里面报告 [0] 0000:00:1d.0 xilinx_aws-vu9p-f1_dynamic_5_0(ts=0xabcd) user(inst=128) WARNING: card(s) marked by '' are not ready, is MPD runing? run 'systemctl status mpd' to check MPD details. ubuntu@:~/aws-fpga$ systemctl status mpd ● mpd.service - Xilinx Management Proxy Daemon (MPD) Loaded: loaded (/etc/systemd/system/mpd.service; disabled; vendor preset: enabled) Active: inactive (dead)

请看还有什么建议？谢谢！

source sdaccel_setup.sh 报错可以忽略，只要AWS_FPGA_REPO_DIR 这个环境变量设置就可以了。
你安装aws包了吗？这个包依赖于xrt包，所以需要在安装xrt包以后安装。安装完aws 包以后，可以检查如下， root@ip-172-31-70-242:~# ls -l /opt/xilinx/xrt/lib/mpd -rw-r--r-- 1 root root 338096 Nov 14 16:59 /opt/xilinx/xrt/lib/libaws_mpd_plugin.so lrwxrwxrwx 1 root root 40 Jan 3 01:01 /opt/xilinx/xrt/lib/libmpd_plugin.so -> /opt/xilinx/xrt/lib/libaws_mpd_plugin.so

查看这个库文件的符号表，应该有很多fpga* 的函数，那些函数来自aws提供的库，表示包编对了。 root@ip-172-31-70-242:~# nm /opt/xilinx/xrt/lib/libaws_mpd_plugin.so |grep fpga

这两个文件是安装aws包时创建的，然后mpd service就运行了。

systemctl status mpd

如果没有运行，手工运行一下，

systemctl start mpd

On Fri, Jan 3, 2020 at 6:16 AM Liang Ma notifications@github.com wrote:

您好：我先不去弄container这些了，先搞AWS和XRT: Follow up 您的建议（AWS support forum 上也给了类似建议，区别只是没有mv XRT XRT_2019.1这一步而已。我的问题好像仍旧..

1.

vi sdaccel_setup.sh 的修改，这一步对我的没有帮助。因为source sdaccel_setup.sh 内部调用执行完sdk_setup.sh PASS 以后报告ERROR: XILINX_SDX variable not set and sdx not in the path--根本没有执行到修改添加的部分就结束退出。这里貌似是因为我没有安装SDAccel开发工具--但是理论上说对于部署来说这个应该不必要对吧？ 2.

后续执行安装完还是老问题，如下：安装过程里面报告

[0] 0000:00:1d.0 xilinx_aws-vu9p-f1_dynamic_5_0(ts=0xabcd) user(inst=128) WARNING: card(s) marked by '' are not ready, is MPD runing? run 'systemctl status mpd' to check MPD details. ubuntu@:~/aws-fpga$ systemctl status mpd ● mpd.service - Xilinx Management Proxy Daemon (MPD) Loaded: loaded (/etc/systemd/system/mpd.service; disabled; vendor preset: enabled) Active: inactive (dead)

请看还有什么建议？谢谢！

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM5ZYGHVXD7FIS57LF3Q35CFPA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEIBGWWI#issuecomment-570583897, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBMZAAQZIH5GRZ4ZDK6TQ35CFPANCNFSM4J26WL3A .

多谢，安装了aws包，mpd service就起来了。不过 xbutil scan 还是报警：现在看还是XCLMGMT没有被安装。我也去XRT 上提了个issue.. $ sudo apt install --reinstall ./xrt_202010.2.5.0_18.04-xrt.deb Reading package lists... Done Building dependency tree Reading state information... Done Note, selecting 'xrt' instead of './xrt_202010.2.5.0_18.04-xrt.deb' 0 upgraded, 0 newly installed, 1 reinstalled, 0 to remove and 0 not upgraded. After this operation, 0 B of additional disk space will be used. Get:1 /home/ubuntu/aws-fpga/SDAccel/Runtime/XRT_2019.1/build/Debug/xrt_202010.2.5.0_18.04-xrt.deb xrt amd64 2.5.0 [44.2 MB] (Reading database ... 132438 files and directories currently installed.) Preparing to unpack .../xrt_202010.2.5.0_18.04-xrt.deb ... Unregistering old XRT Linux kernel module sources 2.5.0 from dkms on Ubuntu/Debian

-------- Uninstall Beginning -------- Module: xrt Version: 2.5.0 Kernel: 4.15.0-1056-aws (x86_64) Status: Before uninstall, this module version was ACTIVE on this kernel.

xocl.ko:

Uninstallation Deleting from: /lib/modules/4.15.0-1056-aws/updates/dkms/ Original module No original module was found for this module on this kernel. Use the dkms install command to reinstall any previous module version. xclmgmt.ko:

DKMS: uninstall completed.

Deleting module version: 2.5.0 completely from the DKMS tree. Done. Cleanup is skipped for package upgrade/downgrade/re-install on Ubuntu/Debian Unpacking xrt (2.5.0) over (2.5.0) ... Setting up xrt (2.5.0) ... Stopping MPD daemon Unloading old XRT Linux kernel modules rmmod: ERROR: Module xclmgmt is not currently loaded Invoking DKMS common.postinst for xrt Loading new xrt-2.5.0 DKMS files... Building for 4.15.0-1056-aws Building initial module for 4.15.0-1056-aws Done.

xocl: Running module version sanity check.

Original module No original module exists within this kernel Installation Installing to /lib/modules/4.15.0-1056-aws/updates/dkms/ xclmgmt.ko: Running module version sanity check.

Original module No original module exists within this kernel Installation Installing to /lib/modules/4.15.0-1056-aws/updates/dkms/ depmod...

DKMS: install completed. Finished DKMS common.postinst Loading new XRT Linux kernel modules modprobe: ERROR: could not insert 'xclmgmt': Unknown symbol in module, or unknown parameter (see dmesg) Installing MSD / MPD daemons Starting MPD daemon Installing pyopencl... ....

$ dmesg | grep xclmgmt [ 699.257353] xclmgmt: Unknown symbol fpga_mgr_unregister (err 0) [ 699.257397] xclmgmt: Unknown symbol fpga_mgr_register (err 0)

XRT Information Version: 2.5.0 Git Hash:
Git Branch: Build Date: 2020-01-03 11:17:41 XOCL: 2.5.0, XCLMGMT: unknown


*[0] 0000:00:1d.0 xilinx_aws-vu9p-f1_dynamic_5_0(ts=0xabcd) user(inst=128)
WARNING: card(s) marked by '*' are not ready, is MPD runing? （MPD 确实已经运行了）

xclmgmt在aws上不需要。xrt重装以后，aws也需要重装。仔细看我前面的回复，告警时检查那两个文件在不在，mpd当前是否运行

On Fri, Jan 3, 2020, 20:37 Liang Ma notifications@github.com wrote:

多谢，安装了aws包，mpd service就起来了。不过 xbutil scan 还是报警：现在看还是XCLMGMT没有被安装。我也去XRT 上提了个issue.. $ sudo apt install --reinstall ./xrt_202010.2.5.0_18.04-xrt.deb Reading package lists... Done Building dependency tree Reading state information... Done Note, selecting 'xrt' instead of './xrt_202010.2.5.0_18.04-xrt.deb' 0 upgraded, 0 newly installed, 1 reinstalled, 0 to remove and 0 not upgraded. After this operation, 0 B of additional disk space will be used. Get:1 /home/ubuntu/aws-fpga/SDAccel/Runtime/XRT_2019.1/build/Debug/xrt_202010.2.5.0_18.04-xrt.deb xrt amd64 2.5.0 [44.2 MB] (Reading database ... 132438 files and directories currently installed.) Preparing to unpack .../xrt_202010.2.5.0_18.04-xrt.deb ... Unregistering old XRT Linux kernel module sources 2.5.0 from dkms on Ubuntu/Debian

-------- Uninstall Beginning -------- Module: xrt Version: 2.5.0 Kernel: 4.15.0-1056-aws (x86_64) Status: Before uninstall, this module version was ACTIVE on this kernel.

xocl.ko:

Uninstallation Deleting from: /lib/modules/4.15.0-1056-aws/updates/dkms/ Original module No original module was found for this module on this kernel. Use the dkms install command to reinstall any previous module version. xclmgmt.ko:

Uninstallation Deleting from: /lib/modules/4.15.0-1056-aws/updates/dkms/ Original module No original module was found for this module on this kernel. Use the dkms install command to reinstall any previous module version. depmod...

DKMS: uninstall completed.

Deleting module version: 2.5.0 completely from the DKMS tree. Done. Cleanup is skipped for package upgrade/downgrade/re-install on Ubuntu/Debian Unpacking xrt (2.5.0) over (2.5.0) ... Setting up xrt (2.5.0) ... Stopping MPD daemon Unloading old XRT Linux kernel modules rmmod: ERROR: Module xclmgmt is not currently loaded Invoking DKMS common.postinst for xrt Loading new xrt-2.5.0 DKMS files... Building for 4.15.0-1056-aws Building initial module for 4.15.0-1056-aws Done.

xocl: Running module version sanity check.

Original module No original module exists within this kernel Installation Installing to /lib/modules/4.15.0-1056-aws/updates/dkms/ xclmgmt.ko: Running module version sanity check.

Original module No original module exists within this kernel Installation Installing to /lib/modules/4.15.0-1056-aws/updates/dkms/ depmod...

DKMS: install completed. Finished DKMS common.postinst Loading new XRT Linux kernel modules modprobe: ERROR: could not insert 'xclmgmt': Unknown symbol in module, or unknown parameter (see dmesg) Installing MSD / MPD daemons Starting MPD daemon Installing pyopencl... ....

$ dmesg | grep xclmgmt [ 699.257353] xclmgmt: Unknown symbol fpga_mgr_unregister (err 0) [ 699.257397] xclmgmt: Unknown symbol fpga_mgr_register (err 0)

XRT Information Version: 2.5.0 Git Hash: Git Branch: Build Date: 2020-01-03 11:17:41 XOCL: 2.5.0, XCLMGMT: unknown

*[0] 0000:00:1d.0 xilinx_aws-vu9p-f1_dynamic_5_0(ts=0xabcd) user(inst=128)

WARNING: card(s) marked by '*' are not ready, is MPD runing? （MPD 确实已经运行了）

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM4PS7ZXDQYQHE4IWNLQ4AHADA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEICQYOI#issuecomment-570756153, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM7QQX5M5VPQQIFMGXTQ4AHADANCNFSM4J26WL3A .

您好，多谢。对了xclmgmt是Xilinx的工具。aws 是fpga-load/clear* 那一套工具是吗？我这台instance上面安装了aws cli，这个会有冲突吗？

我在 reinstall ×-xrt.deb包以后，也重新reinstall了 *-aws.deb包。您给的几个建议都检查了，都是正常的。也尝试了一次reboot。但是warning还是依旧。。如下：

$ xbutil validate INFO: Found 1 cards INFO: Validating card[0]: xilinx_aws-vu9p-f1_dynamic_5_0 <......> ERROR: Failed to find xclbin in /opt/xilinx/xsa/xilinx_aws-vu9p-f1_dynamic_5_0/test/ and /opt/xilinx/dsa/xilinx_aws-vu9p-f1_dynamic_5_0/test/ERROR: Failed to find 22_verify.py or verify.xclbin, Shell package not installed properly. ERROR: == verify kernel test FAILED INFO: Card[0] failed to validate. ERROR: Some cards failed to validate. $ xbutil scan INFO: Found total 1 card(s), 0 are usable <......> [0] 0000:00:1d.0 xilinx_aws-vu9p-f1_dynamic_5_0(ts=0xabcd) user(inst=128) WARNING: card(s) marked by '' are not ready, is MPD runing? run 'systemctl status mpd' to check MPD details. $ systemctl status mpd ● mpd.service - Xilinx Management Proxy Daemon (MPD) Loaded: loaded (/etc/systemd/system/mpd.service; enabled; vendor preset: enabled) Active: active (running) since Sat 2020-01-04 08:17:08 UTC; 3min 8s ago Main PID: 17775 (mpd) Tasks: 3 (limit: 4915) CGroup: /system.slice/mpd.service └─17775 /opt/xilinx/xrt/bin/mpd

<.....> $ nm /opt/xilinx/xrt/lib/libaws_mpd_plugin.so | grep fpga 000000000005ce1c T _ZN6AwsDev18checkAndSkipReloadEPcP20fpga_mgmt_image_info 00000000000996bc T fpga_acquire_readdir_lock 000000000009b893 T fpga_hal_mbox_attach <.....> $ ls -l /opt/xilinx/xrt/lib/mpd -rw-r--r-- 1 root root 2537760 Jan 3 12:58 /opt/xilinx/xrt/lib/libaws_mpd_plugin.so lrwxrwxrwx 1 root root 40 Jan 4 08:17 /opt/xilinx/xrt/lib/libmpd_plugin.so -> /opt/xilinx/xrt/lib/libaws_mpd_plugin.so

aws CLI 和mpd 都调用同一个库，不会冲突。另外validate不能在aws上运行。建议是开着log，tail -f /var/log/syslog，然后另一个窗口systemctl restart mpd，xbutil scan，看看log里有什么可疑的，或者你把log文件发给我看看。你这是最新的xrt，说不定有什么问题，2019.2 分支是发行版，比较稳定

On Sat, Jan 4, 2020, 00:34 Liang Ma notifications@github.com wrote:

您好，多谢。对了xclmgmt是Xilinx的工具。aws 是fpga-load/clear* 那一套工具是吗？我这台instance上面安装了aws cli，这个会有冲突吗？

我在 reinstall ×-xrt.deb包以后，也重新reinstall了 *-aws.deb包。您给的几个建议都检查了，都是正常的。也尝试了一次reboot。但是warning还是依旧。。如下：

$ xbutil validate INFO: Found 1 cards INFO: Validating card[0]: xilinx_aws-vu9p-f1_dynamic_5_0 <......> ERROR: Failed to find xclbin in /opt/xilinx/xsa/xilinx_aws-vu9p-f1_dynamic_5_0/test/ and /opt/xilinx/dsa/xilinx_aws-vu9p-f1_dynamic_5_0/test/ERROR: Failed to find 22_verify.py or verify.xclbin, Shell package not installed properly. ERROR: == verify kernel test FAILED INFO: Card[0] failed to validate. ERROR: Some cards failed to validate. $ xbutil scan INFO: Found total 1 card(s), 0 are usable <......>

[0] 0000:00:1d.0 xilinx_aws-vu9p-f1_dynamic_5_0(ts=0xabcd) user(inst=128) WARNING: card(s) marked by '' are not ready, is MPD runing? run 'systemctl status mpd' to check MPD details. $ systemctl status mpd ● mpd.service - Xilinx Management Proxy Daemon (MPD) Loaded: loaded (/etc/systemd/system/mpd.service; enabled; vendor preset: enabled) Active: active (running) since Sat 2020-01-04 08:17:08 UTC; 3min 8s ago Main PID: 17775 (mpd) Tasks: 3 (limit: 4915) CGroup: /system.slice/mpd.service └─17775 /opt/xilinx/xrt/bin/mpd

<.....> $ nm /opt/xilinx/xrt/lib/libaws_mpd_plugin.so | grep fpga 000000000005ce1c T _ZN6AwsDev18checkAndSkipReloadEPcP20fpga_mgmt_image_info 00000000000996bc T fpga_acquire_readdir_lock 000000000009b893 T fpga_hal_mbox_attach <.....> $ ls -l /opt/xilinx/xrt/lib/mpd -rw-r--r-- 1 root root 2537760 Jan 3 12:58 /opt/xilinx/xrt/lib/libaws_mpd_plugin.so lrwxrwxrwx 1 root root 40 Jan 4 08:17 /opt/xilinx/xrt/lib/libmpd_plugin.so -> /opt/xilinx/xrt/lib/libaws_mpd_plugin.so

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM7FF5KPXVNEGPST2YTQ4BCX3A5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEICTRAA#issuecomment-570767488, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBM2JBPP3F5SEHQMGH2LQ4BCX3ANCNFSM4J26WL3A .

如果要退回到2019.2发行版，请问如何操作？如下是log，请帮忙看一下--我没看出有什么特别的。。


~$ tail -f /var/log/syslog
Jan  6 03:39:30 ip systemd[23845]: Listening on REST API socket for snapd user session agent.
Jan  6 03:39:30 ip systemd[23845]: Reached target Timers.
Jan  6 03:39:30 ip systemd[23845]: Listening on GnuPG cryptographic agent and passphrase cache (access for web browsers).
Jan  6 03:39:30 ip systemd[23845]: Listening on GnuPG cryptographic agent (ssh-agent emulation).
Jan  6 03:39:30 ip systemd[23845]: Reached target Sockets.
Jan  6 03:39:30 ip systemd[23845]: Reached target Basic System.
Jan  6 03:39:30 ip systemd[23845]: Reached target Default.
Jan  6 03:39:30 ip systemd[23845]: Startup finished in 17ms.
Jan  6 03:39:30 ip systemd[1]: Started User Manager for UID 1000.
Jan  6 03:40:34 ip systemd[1]: Started Session 65 of user ubuntu.
Jan  6 03:42:18 ip mpd[17775]: mpd caught signal 15
Jan  6 03:42:18 ip systemd[1]: Stopping Xilinx Management Proxy Daemon (MPD)...
Jan  6 03:42:18 ip mpd[17775]: 0000:00:1d.0 handleMsg thread exit
Jan  6 03:42:19 ip mpd[17775]: [0:0:1d.0] write 56 bytes out of 56 bytes to fd 4
Jan  6 03:42:19 ip mpd[17775]: [0:0:1d.0] mpd_getMsg thread 0 exit!!
Jan  6 03:42:19 ip kernel: [171301.881911] mailbox.u mailbox.u.15728640: dev ffff98f8fb9acc10, xocl_drvinst_close: CLOSE 2
Jan  6 03:42:19 ip kernel: [171301.881915] mailbox.u mailbox.u.15728640: dev ffff98f8fb9acc10, xocl_drvinst_close: NOTIFY 0000000047f37cb6
Jan  6 03:42:19 ip kernel: [171301.882040] xocl 0000:00:1d.0: dev ffff98f8f8c5a0a0, __xocl_subdev_destroy: Destroy subdev mb_scheduler, cdev           (null)
Jan  6 03:42:19 ip kernel: [171301.882085] [drm] /var/lib/dkms/xrt/2.5.0/build/driver/xocl/userpf/../subdev/mb_scheduler.c:3495 scheduler thread exits with value 0
Jan  6 03:42:19 ip kernel: [171301.882135] [drm] command scheduler removed
Jan  6 03:42:19 ip kernel: [171301.882167] xocl 0000:00:1d.0: dev ffff98f8f8c5a0a0, __xocl_subdev_destroy: Destroy subdev dma.xdma, cdev           (null)
Jan  6 03:42:19 ip kernel: [171301.909038] xocl 0000:00:1d.0: dev ffff98f8f8c5a0a0, __xocl_subdev_destroy: Destroy subdev rom, cdev           (null)
Jan  6 03:42:19 ip kernel: [171301.909048] rom.u rom.u.0: dev ffff98f8fa20f010, feature_rom_remove: Remove feature rom
Jan  6 03:42:22 ip kernel: [171304.911975] pci 0000:00:1d.0: [1d0f:1042] type 00 class 0x058000
Jan  6 03:42:22 ip kernel: [171304.912594] pci 0000:00:1d.0: reg 0x10: [mem 0x82000000-0x83ffffff]
Jan  6 03:42:22 ip kernel: [171304.913005] pci 0000:00:1d.0: reg 0x14: [mem 0x85400000-0x855fffff]
Jan  6 03:42:22 ip kernel: [171304.913598] pci 0000:00:1d.0: reg 0x18: [mem 0x85600000-0x8560ffff 64bit pref]
Jan  6 03:42:22 ip kernel: [171304.914134] pci 0000:00:1d.0: reg 0x20: [mem 0x2000000000-0x3fffffffff 64bit pref]
Jan  6 03:42:22 ip kernel: [171304.917466] pci 0000:00:1d.0: BAR 4: assigned [mem 0x2000000000-0x3fffffffff 64bit pref]
Jan  6 03:42:22 ip kernel: [171304.917808] pci 0000:00:1d.0: BAR 0: assigned [mem 0x82000000-0x83ffffff]
Jan  6 03:42:22 ip kernel: [171304.917908] pci 0000:00:1d.0: BAR 1: assigned [mem 0x85400000-0x855fffff]
Jan  6 03:42:22 ip kernel: [171304.918014] pci 0000:00:1d.0: BAR 2: assigned [mem 0x85600000-0x8560ffff 64bit pref]
Jan  6 03:42:22 ip kernel: [171304.918767] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_subdev_vsec: No Vendor Specific Capability.
Jan  6 03:42:22 ip kernel: [171304.921587] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev rom.u
Jan  6 03:42:22 ip kernel: [171304.921591] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: resource [mem 0x820b0000-0x820b0fff]
Jan  6 03:42:22 ip kernel: [171304.921646] rom.u rom.u.0: dev ffff98f8f9eb2410, get_header_from_iomem: Found AWS VU9P Device without featureROM
Jan  6 03:42:22 ip kernel: [171304.921647] rom.u rom.u.0: dev ffff98f8f9eb2410, get_header_from_iomem: Enabling AWS dynamic 5.0 Shell
Jan  6 03:42:22 ip kernel: [171304.921655] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: ROM magic : xlnx
Jan  6 03:42:22 ip kernel: [171304.921656] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: VBNV: xilinx_aws-vu9p-f1_dynamic_5_0
Jan  6 03:42:22 ip kernel: [171304.921657] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: DDR channel count : 4
Jan  6 03:42:22 ip kernel: [171304.921659] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: DDR channel size: 16 GB
Jan  6 03:42:22 ip kernel: [171304.921660] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: Major Version: 4
Jan  6 03:42:22 ip kernel: [171304.921661] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: Minor Version: 0
Jan  6 03:42:22 ip kernel: [171304.921662] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: IPBuildID: 43981
Jan  6 03:42:22 ip kernel: [171304.921663] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: TimeSinceEpoch: abcd
Jan  6 03:42:22 ip kernel: [171304.921664] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: FeatureBitMap: 1
Jan  6 03:42:22 ip kernel: [171304.921683] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev rom inst 0
Jan  6 03:42:22 ip kernel: [171304.921685] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev rom.u inst 0 is active
Jan  6 03:42:22 ip kernel: [171304.921698] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev rom.u
Jan  6 03:42:22 ip kernel: [171304.921699] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev dma.xdma.u
Jan  6 03:42:22 ip kernel: [171304.921727] xocl:xdma_device_open: xocl device 0000:00:1d.0, 0x00000000ca0f8fd7.
Jan  6 03:42:22 ip kernel: [171304.922066] xocl:map_single_bar: map bar 0, len 33554432
Jan  6 03:42:22 ip kernel: [171304.922068] xocl:map_single_bar: map bar 1, len 2097152
Jan  6 03:42:22 ip kernel: [171304.922074] xocl:map_single_bar: BAR1 at 0x85400000 mapped at 0x00000000958b4a7a, length=2097152(/2097152)
Jan  6 03:42:22 ip kernel: [171304.922077] xocl:map_single_bar: map bar 2, len 65536
Jan  6 03:42:22 ip kernel: [171304.922093] xocl:map_single_bar: BAR2 at 0x85600000 mapped at 0x000000001fecee6f, length=65536(/65536)
Jan  6 03:42:22 ip kernel: [171304.922095] xocl:map_bars: config bar 2, pos 1.
Jan  6 03:42:22 ip kernel: [171304.922096] xocl:map_single_bar: map bar 3, len 0
Jan  6 03:42:22 ip kernel: [171304.922097] xocl:map_single_bar: map bar 4, len 137438953472
Jan  6 03:42:22 ip kernel: [171304.922098] xocl:map_single_bar: map bar 5, len 0
Jan  6 03:42:22 ip kernel: [171304.922098] xocl:identify_bars: xdev 0x0000000089fb3b0d, bars 2, config at 1.
Jan  6 03:42:22 ip kernel: [171304.922100] xocl:identify_bars: 2 BARs: config 1, user 1, bypass -1.
Jan  6 03:42:22 ip kernel: [171304.928989] xocl:enable_msi_msix: request vectors: h2c 4, c2h 4, user 16
Jan  6 03:42:22 ip kernel: [171304.929044] xocl:pci_keep_intx_enabled: 0000:00:1d.0: clear INTX_DISABLE, 0x406 -> 0x6.
Jan  6 03:42:22 ip kernel: [171304.929124] xocl:irq_msix_channel_setup: engine 0-H2C0-MM, irq#122.
Jan  6 03:42:22 ip kernel: [171304.929144] xocl:irq_msix_channel_setup: engine 0-H2C1-MM, irq#123.
Jan  6 03:42:22 ip kernel: [171304.929159] xocl:irq_msix_channel_setup: engine 0-H2C2-MM, irq#124.
Jan  6 03:42:22 ip kernel: [171304.929177] xocl:irq_msix_channel_setup: engine 0-H2C3-MM, irq#125.
Jan  6 03:42:22 ip kernel: [171304.929194] xocl:irq_msix_channel_setup: engine 0-C2H0-MM, irq#126.
Jan  6 03:42:22 ip kernel: [171304.929210] xocl:irq_msix_channel_setup: engine 0-C2H1-MM, irq#127.
Jan  6 03:42:22 ip kernel: [171304.929228] xocl:irq_msix_channel_setup: engine 0-C2H2-MM, irq#128.
Jan  6 03:42:22 ip kernel: [171304.929243] xocl:irq_msix_channel_setup: engine 0-C2H3-MM, irq#129.
Jan  6 03:42:22 ip kernel: [171304.929264] xocl:irq_msix_user_setup: 0-USR-0, IRQ#130 with 0x00000000826da15c
Jan  6 03:42:22 ip kernel: [171304.929280] xocl:irq_msix_user_setup: 0-USR-1, IRQ#131 with 0x000000003fcd4350
Jan  6 03:42:22 ip kernel: [171304.929299] xocl:irq_msix_user_setup: 0-USR-2, IRQ#132 with 0x00000000daa9e757
Jan  6 03:42:22 ip kernel: [171304.929319] xocl:irq_msix_user_setup: 0-USR-3, IRQ#133 with 0x000000007697b51b
Jan  6 03:42:22 ip kernel: [171304.929335] xocl:irq_msix_user_setup: 0-USR-4, IRQ#134 with 0x000000000c6e7cfd
Jan  6 03:42:22 ip kernel: [171304.929353] xocl:irq_msix_user_setup: 0-USR-5, IRQ#135 with 0x00000000357f26e1
Jan  6 03:42:22 ip kernel: [171304.929370] xocl:irq_msix_user_setup: 0-USR-6, IRQ#136 with 0x0000000002013885
Jan  6 03:42:22 ip kernel: [171304.929388] xocl:irq_msix_user_setup: 0-USR-7, IRQ#137 with 0x00000000a00937ae
Jan  6 03:42:22 ip kernel: [171304.929404] xocl:irq_msix_user_setup: 0-USR-8, IRQ#138 with 0x00000000e89d8347
Jan  6 03:42:22 ip kernel: [171304.929421] xocl:irq_msix_user_setup: 0-USR-9, IRQ#139 with 0x000000003cafb0c5
Jan  6 03:42:22 ip kernel: [171304.929438] xocl:irq_msix_user_setup: 0-USR-10, IRQ#140 with 0x00000000c8eac97e
Jan  6 03:42:22 ip kernel: [171304.929454] xocl:irq_msix_user_setup: 0-USR-11, IRQ#141 with 0x0000000083301800
Jan  6 03:42:22 ip kernel: [171304.929472] xocl:irq_msix_user_setup: 0-USR-12, IRQ#142 with 0x00000000b95d63c4
Jan  6 03:42:22 ip kernel: [171304.929488] xocl:irq_msix_user_setup: 0-USR-13, IRQ#143 with 0x000000005ab1eb56
Jan  6 03:42:22 ip kernel: [171304.929506] xocl:irq_msix_user_setup: 0-USR-14, IRQ#144 with 0x0000000011ac446a
Jan  6 03:42:22 ip kernel: [171304.929525] xocl:irq_msix_user_setup: 0-USR-15, IRQ#145 with 0x00000000f74a36ff
Jan  6 03:42:22 ip kernel: [171304.929561] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev dma.xdma inst 2097152
Jan  6 03:42:22 ip kernel: [171304.929563] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev dma.xdma.u inst 2097152 is active
Jan  6 03:42:22 ip kernel: [171304.929565] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev mb_scheduler.u
Jan  6 03:42:22 ip kernel: [171304.929569] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: resource [mem 0x82180000-0x82180fff]
Jan  6 03:42:22 ip kernel: [171304.929570] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: resource [mem 0x82190000-0x8219ffff]
Jan  6 03:42:22 ip kernel: [171304.929572] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: resource [irq 0-3]
Jan  6 03:42:22 ip kernel: [171304.929632] xocl_mb_sche mb_scheduler.u.5242880: dev ffff98f8f9eb0810, exec_create: CQ size is 65536
Jan  6 03:42:22 ip kernel: [171304.929643] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, exec_reset: exec_reset(2) cfg(0)
Jan  6 03:42:22 ip kernel: [171304.929645] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, exec_reset: exec_reset resets
Jan  6 03:42:22 ip kernel: [171304.929647] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, exec_reset: exec->xclbin(00000000-0000-0000-0000-000000000000),xclbin(00000000-0000-0000-0000-000000000000)
Jan  6 03:42:22 ip kernel: [171304.929753] [drm] command scheduler started
Jan  6 03:42:22 ip kernel: [171304.929767] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev mb_scheduler inst 5242880
Jan  6 03:42:22 ip kernel: [171304.929769] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev mb_scheduler.u inst 5242880 is active
Jan  6 03:42:22 ip kernel: [171304.929770] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev mailbox.u
Jan  6 03:42:22 ip kernel: [171304.930038] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_probe: successfully initialized
Jan  6 03:42:22 ip kernel: [171304.930052] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev mailbox inst 15728640
Jan  6 03:42:22 ip kernel: [171304.930139] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev mailbox.u inst 15728640 is active
Jan  6 03:42:22 ip kernel: [171304.930140] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev icap.u
Jan  6 03:42:22 ip kernel: [171304.930172] icap.u icap.u.16777216: dev ffff98f8fa20ac10, icap_probe: successfully initialized FPGA IDCODE 0x0
Jan  6 03:42:22 ip kernel: [171304.930182] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev icap inst 16777216
Jan  6 03:42:22 ip kernel: [171304.930183] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev icap.u inst 16777216 is active
Jan  6 03:42:22 ip kernel: [171304.930186] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_subdev_vsec: No Vendor Specific Capability.
Jan  6 03:42:22 ip kernel: [171304.930187] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_subdev_vsec: No Vendor Specific Capability.
Jan  6 03:42:22 ip kernel: [171304.930190] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_p2p_init: Initializing P2P, bar 4, len 137438953472
Jan  6 03:42:22 ip kernel: [171304.930247] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 15 via SW
Jan  6 03:42:25 ip mpd[17775]: [0:0:1d.0] write 0 bytes out of 2104 bytes to fd 4
Jan  6 03:42:25 ip mpd[17775]: [0:0:1d.0] mpd_handleMsg thread 0 exit!!
Jan  6 03:42:25 ip mpd[17775]: 0000:00:1d.0 getMsg thread exit
Jan  6 03:42:25 ip mpd[17775]: aws mpd plugin fini called
Jan  6 03:42:25 ip mpd[17775]: ended
Jan  6 03:42:25 ip systemd[1]: Stopped Xilinx Management Proxy Daemon (MPD).
Jan  6 03:42:25 ip kernel: [171307.352907] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, timeout_msg: found outstanding msg time'd out
Jan  6 03:42:25 ip kernel: [171307.352910] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, timeout_msg: peer becomes dead
Jan  6 03:42:25 ip kernel: [171307.352964] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62
Jan  6 03:42:25 ip kernel: [171307.353438] [drm] Initialized xocl 2.5.0 20200103 for 0000:00:1d.0 on minor 0
Jan  6 03:42:25 ip kernel: [171307.353505] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 11 via SW
Jan  6 03:42:25 ip kernel: [171307.353508] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mb_connect: ch_state 0x0, ret -107
Jan  6 03:42:25 ip systemd[1]: Started Xilinx Management Proxy Daemon (MPD).
Jan  6 03:42:25 ip mpd[24133]: started
Jan  6 03:42:25 ip mpd[24133]: found mpd plugin: /opt/xilinx/xrt/lib/libmpd_plugin.so
Jan  6 03:42:25 ip mpd[24133]: aws mpd plugin init called: 0
Jan  6 03:42:25 ip mpd[24133]: create thread pair for 0000:00:1d.0
Jan  6 03:42:25 ip mpd[24133]: 1 pairs of threads running...
Jan  6 03:42:25 ip mpd[24133]: [0:0:1d.0] write 56 bytes out of 56 bytes to fd 4
Jan  6 03:42:25 ip mpd[24133]: [0:0:1d.0] msg arrived on mailbox fd 4
Jan  6 03:42:25 ip mpd[24133]: [0:0:1d.0] retrieved msg size from mailbox: 40 bytes
Jan  6 03:42:25 ip mpd[24133]: [0:0:1d.0] read 72 bytes out of 72 bytes from fd 4, valid: 1
Jan  6 03:42:25 ip mpd[24133]: [0:0:1d.0] mpd daemon: request 11 received(reqSize: 24)
Jan  6 03:42:25 ip kernel: [171307.363135] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, _xocl_drvinst_open: OPEN 1
Jan  6 03:42:25 ip kernel: [171307.363214] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, dequeue_rx_msg: peer becomes active
Jan  6 03:42:25 ip kernel: [171307.363249] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, process_request: received request from peer: 12, passed on
Jan  6 03:42:25 ip kernel: [171307.363252] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mailbox_srv: received request (12) from peer
Jan  6 03:42:25 ip kernel: [171307.363253] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mailbox_srv: mgmt driver online
Jan  6 03:42:25 ip kernel: [171307.363286] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 11 via SW
Jan  6 03:42:25 ip kernel: [171307.363384] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_read: Software TX msg is too big
Jan  6 03:42:25 ip kernel: [171307.364904] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_refresh_subdevs: get fdt from peer
Jan  6 03:42:25 ip mpd[24133]: [0:0:1d.0] msg arrived on mailbox fd 4
Jan  6 03:42:25 ip mpd[24133]: [0:0:1d.0] retrieved msg size from mailbox: 48 bytes
Jan  6 03:42:25 ip mpd[24133]: [0:0:1d.0] read 80 bytes out of 80 bytes from fd 4, valid: 1
Jan  6 03:42:25 ip kernel: [171307.365337] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 10 via SW
Jan  6 03:42:25 ip kernel: [171307.365436] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_read: Software TX msg is too big
Jan  6 03:42:49 ip mpd[24133]: [0:0:1d.0] msg arrived on mailbox fd 4
Jan  6 03:42:49 ip mpd[24133]: [0:0:1d.0] retrieved msg size from mailbox: 32 bytes
Jan  6 03:42:49 ip mpd[24133]: [0:0:1d.0] read 64 bytes out of 64 bytes from fd 4, valid: 1
Jan  6 03:42:49 ip kernel: [171331.388847] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, timeout_msg: found waiting msg time'd out
Jan  6 03:42:49 ip kernel: [171331.397224] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mb_connect: ch_state 0x0, ret -62
Jan  6 03:42:49 ip kernel: [171331.397457] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 15 via SW
Jan  6 03:42:49 ip kernel: [171331.397606] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_read: Software TX msg is too big
Jan  6 03:43:13 ip kernel: [171355.428756] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, timeout_msg: found waiting msg time'd out
Jan  6 03:43:13 ip kernel: [171355.435262] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62
Jan  6 03:43:13 ip kernel: [171355.435273] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_destroy: Destroy subdev icap, cdev           (null)
Jan  6 03:43:13 ip kernel: [171355.435292] icap.u icap.u.16777216: dev ffff98f8fa20ac10, icap_remove: cleaned up successfully
Jan  6 03:43:13 ip kernel: [171355.435338] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_destroy: Destroy subdev mailbox, cdev 00000000db05ef2b
Jan  6 03:43:13 ip kernel: [171355.435588] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_remove: mailbox cleaned up successfully
Jan  6 03:43:13 ip kernel: [171355.435592] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, xocl_drvinst_free: Wait for close 00000000726b68db

不知道这一条是因为什么出现的，

Jan 6 03:42:25 ip mpd[17775]: [0:0:1d.0] write 0 bytes out of 2104 bytes to fd 4

总之，出现后daemon就退出了，mailbox 信息发不了就timeout了。但是之后daemon重启应该是可以恢复的，不知道为什么超时的消息没有删除。

你试试以下命令，再看看log，

systemctl stop mpd

rmmod xocl

modprobe xocl

systemctl start mpd

如果用2019.2，在clone xrt时加上 '-b 2019.2'

On Sun, Jan 5, 2020 at 7:47 PM Liang Ma notifications@github.com wrote:

如果要退回到2019.2发行版，请问如何操作？如下是log，请帮忙看一下--我没看出有什么特别的。。

~$ tail -f /var/log/syslog

Jan 6 03:39:30 ip systemd[23845]: Listening on REST API socket for snapd user session agent.

Jan 6 03:39:30 ip systemd[23845]: Reached target Timers.

Jan 6 03:39:30 ip systemd[23845]: Listening on GnuPG cryptographic agent and passphrase cache (access for web browsers).

Jan 6 03:39:30 ip systemd[23845]: Listening on GnuPG cryptographic agent (ssh-agent emulation).

Jan 6 03:39:30 ip systemd[23845]: Reached target Sockets.

Jan 6 03:39:30 ip systemd[23845]: Reached target Basic System.

Jan 6 03:39:30 ip systemd[23845]: Reached target Default.

Jan 6 03:39:30 ip systemd[23845]: Startup finished in 17ms.

Jan 6 03:39:30 ip systemd[1]: Started User Manager for UID 1000.

Jan 6 03:40:34 ip systemd[1]: Started Session 65 of user ubuntu.

Jan 6 03:42:18 ip mpd[17775]: mpd caught signal 15

Jan 6 03:42:18 ip systemd[1]: Stopping Xilinx Management Proxy Daemon (MPD)...

Jan 6 03:42:18 ip mpd[17775]: 0000:00:1d.0 handleMsg thread exit

Jan 6 03:42:19 ip mpd[17775]: [0:0:1d.0] write 56 bytes out of 56 bytes to fd 4

Jan 6 03:42:19 ip mpd[17775]: [0:0:1d.0] mpd_getMsg thread 0 exit!!

Jan 6 03:42:19 ip kernel: [171301.881911] mailbox.u mailbox.u.15728640: dev ffff98f8fb9acc10, xocl_drvinst_close: CLOSE 2

Jan 6 03:42:19 ip kernel: [171301.881915] mailbox.u mailbox.u.15728640: dev ffff98f8fb9acc10, xocl_drvinst_close: NOTIFY 0000000047f37cb6

Jan 6 03:42:19 ip kernel: [171301.882040] xocl 0000:00:1d.0: dev ffff98f8f8c5a0a0, __xocl_subdev_destroy: Destroy subdev mb_scheduler, cdev (null)

Jan 6 03:42:19 ip kernel: [171301.882085] [drm] /var/lib/dkms/xrt/2.5.0/build/driver/xocl/userpf/../subdev/mb_scheduler.c:3495 scheduler thread exits with value 0

Jan 6 03:42:19 ip kernel: [171301.882135] [drm] command scheduler removed

Jan 6 03:42:19 ip kernel: [171301.882167] xocl 0000:00:1d.0: dev ffff98f8f8c5a0a0, __xocl_subdev_destroy: Destroy subdev dma.xdma, cdev (null)

Jan 6 03:42:19 ip kernel: [171301.909038] xocl 0000:00:1d.0: dev ffff98f8f8c5a0a0, __xocl_subdev_destroy: Destroy subdev rom, cdev (null)

Jan 6 03:42:19 ip kernel: [171301.909048] rom.u rom.u.0: dev ffff98f8fa20f010, feature_rom_remove: Remove feature rom

Jan 6 03:42:22 ip kernel: [171304.911975] pci 0000:00:1d.0: [1d0f:1042] type 00 class 0x058000

Jan 6 03:42:22 ip kernel: [171304.912594] pci 0000:00:1d.0: reg 0x10: [mem 0x82000000-0x83ffffff]

Jan 6 03:42:22 ip kernel: [171304.913005] pci 0000:00:1d.0: reg 0x14: [mem 0x85400000-0x855fffff]

Jan 6 03:42:22 ip kernel: [171304.913598] pci 0000:00:1d.0: reg 0x18: [mem 0x85600000-0x8560ffff 64bit pref]

Jan 6 03:42:22 ip kernel: [171304.914134] pci 0000:00:1d.0: reg 0x20: [mem 0x2000000000-0x3fffffffff 64bit pref]

Jan 6 03:42:22 ip kernel: [171304.917466] pci 0000:00:1d.0: BAR 4: assigned [mem 0x2000000000-0x3fffffffff 64bit pref]

Jan 6 03:42:22 ip kernel: [171304.917808] pci 0000:00:1d.0: BAR 0: assigned [mem 0x82000000-0x83ffffff]

Jan 6 03:42:22 ip kernel: [171304.917908] pci 0000:00:1d.0: BAR 1: assigned [mem 0x85400000-0x855fffff]

Jan 6 03:42:22 ip kernel: [171304.918014] pci 0000:00:1d.0: BAR 2: assigned [mem 0x85600000-0x8560ffff 64bit pref]

Jan 6 03:42:22 ip kernel: [171304.918767] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_subdev_vsec: No Vendor Specific Capability.

Jan 6 03:42:22 ip kernel: [171304.921587] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev rom.u

Jan 6 03:42:22 ip kernel: [171304.921591] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: resource [mem 0x820b0000-0x820b0fff]

Jan 6 03:42:22 ip kernel: [171304.921646] rom.u rom.u.0: dev ffff98f8f9eb2410, get_header_from_iomem: Found AWS VU9P Device without featureROM

Jan 6 03:42:22 ip kernel: [171304.921647] rom.u rom.u.0: dev ffff98f8f9eb2410, get_header_from_iomem: Enabling AWS dynamic 5.0 Shell

Jan 6 03:42:22 ip kernel: [171304.921655] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: ROM magic : xlnx

Jan 6 03:42:22 ip kernel: [171304.921656] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: VBNV: xilinx_aws-vu9p-f1_dynamic_5_0

Jan 6 03:42:22 ip kernel: [171304.921657] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: DDR channel count : 4

Jan 6 03:42:22 ip kernel: [171304.921659] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: DDR channel size: 16 GB

Jan 6 03:42:22 ip kernel: [171304.921660] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: Major Version: 4

Jan 6 03:42:22 ip kernel: [171304.921661] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: Minor Version: 0

Jan 6 03:42:22 ip kernel: [171304.921662] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: IPBuildID: 43981

Jan 6 03:42:22 ip kernel: [171304.921663] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: TimeSinceEpoch: abcd

Jan 6 03:42:22 ip kernel: [171304.921664] rom.u rom.u.0: dev ffff98f8f9eb2410, feature_rom_probe: FeatureBitMap: 1

Jan 6 03:42:22 ip kernel: [171304.921683] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev rom inst 0

Jan 6 03:42:22 ip kernel: [171304.921685] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev rom.u inst 0 is active

Jan 6 03:42:22 ip kernel: [171304.921698] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev rom.u

Jan 6 03:42:22 ip kernel: [171304.921699] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev dma.xdma.u

Jan 6 03:42:22 ip kernel: [171304.921727] xocl:xdma_device_open: xocl device 0000:00:1d.0, 0x00000000ca0f8fd7.

Jan 6 03:42:22 ip kernel: [171304.922066] xocl:map_single_bar: map bar 0, len 33554432

Jan 6 03:42:22 ip kernel: [171304.922068] xocl:map_single_bar: map bar 1, len 2097152

Jan 6 03:42:22 ip kernel: [171304.922074] xocl:map_single_bar: BAR1 at 0x85400000 mapped at 0x00000000958b4a7a, length=2097152(/2097152)

Jan 6 03:42:22 ip kernel: [171304.922077] xocl:map_single_bar: map bar 2, len 65536

Jan 6 03:42:22 ip kernel: [171304.922093] xocl:map_single_bar: BAR2 at 0x85600000 mapped at 0x000000001fecee6f, length=65536(/65536)

Jan 6 03:42:22 ip kernel: [171304.922095] xocl:map_bars: config bar 2, pos 1.

Jan 6 03:42:22 ip kernel: [171304.922096] xocl:map_single_bar: map bar 3, len 0

Jan 6 03:42:22 ip kernel: [171304.922097] xocl:map_single_bar: map bar 4, len 137438953472

Jan 6 03:42:22 ip kernel: [171304.922098] xocl:map_single_bar: map bar 5, len 0

Jan 6 03:42:22 ip kernel: [171304.922098] xocl:identify_bars: xdev 0x0000000089fb3b0d, bars 2, config at 1.

Jan 6 03:42:22 ip kernel: [171304.922100] xocl:identify_bars: 2 BARs: config 1, user 1, bypass -1.

Jan 6 03:42:22 ip kernel: [171304.928989] xocl:enable_msi_msix: request vectors: h2c 4, c2h 4, user 16

Jan 6 03:42:22 ip kernel: [171304.929044] xocl:pci_keep_intx_enabled: 0000:00:1d.0: clear INTX_DISABLE, 0x406 -> 0x6.

Jan 6 03:42:22 ip kernel: [171304.929124] xocl:irq_msix_channel_setup: engine 0-H2C0-MM, irq#122.

Jan 6 03:42:22 ip kernel: [171304.929144] xocl:irq_msix_channel_setup: engine 0-H2C1-MM, irq#123.

Jan 6 03:42:22 ip kernel: [171304.929159] xocl:irq_msix_channel_setup: engine 0-H2C2-MM, irq#124.

Jan 6 03:42:22 ip kernel: [171304.929177] xocl:irq_msix_channel_setup: engine 0-H2C3-MM, irq#125.

Jan 6 03:42:22 ip kernel: [171304.929194] xocl:irq_msix_channel_setup: engine 0-C2H0-MM, irq#126.

Jan 6 03:42:22 ip kernel: [171304.929210] xocl:irq_msix_channel_setup: engine 0-C2H1-MM, irq#127.

Jan 6 03:42:22 ip kernel: [171304.929228] xocl:irq_msix_channel_setup: engine 0-C2H2-MM, irq#128.

Jan 6 03:42:22 ip kernel: [171304.929243] xocl:irq_msix_channel_setup: engine 0-C2H3-MM, irq#129.

Jan 6 03:42:22 ip kernel: [171304.929264] xocl:irq_msix_user_setup: 0-USR-0, IRQ#130 with 0x00000000826da15c

Jan 6 03:42:22 ip kernel: [171304.929280] xocl:irq_msix_user_setup: 0-USR-1, IRQ#131 with 0x000000003fcd4350

Jan 6 03:42:22 ip kernel: [171304.929299] xocl:irq_msix_user_setup: 0-USR-2, IRQ#132 with 0x00000000daa9e757

Jan 6 03:42:22 ip kernel: [171304.929319] xocl:irq_msix_user_setup: 0-USR-3, IRQ#133 with 0x000000007697b51b

Jan 6 03:42:22 ip kernel: [171304.929335] xocl:irq_msix_user_setup: 0-USR-4, IRQ#134 with 0x000000000c6e7cfd

Jan 6 03:42:22 ip kernel: [171304.929353] xocl:irq_msix_user_setup: 0-USR-5, IRQ#135 with 0x00000000357f26e1

Jan 6 03:42:22 ip kernel: [171304.929370] xocl:irq_msix_user_setup: 0-USR-6, IRQ#136 with 0x0000000002013885

Jan 6 03:42:22 ip kernel: [171304.929388] xocl:irq_msix_user_setup: 0-USR-7, IRQ#137 with 0x00000000a00937ae

Jan 6 03:42:22 ip kernel: [171304.929404] xocl:irq_msix_user_setup: 0-USR-8, IRQ#138 with 0x00000000e89d8347

Jan 6 03:42:22 ip kernel: [171304.929421] xocl:irq_msix_user_setup: 0-USR-9, IRQ#139 with 0x000000003cafb0c5

Jan 6 03:42:22 ip kernel: [171304.929438] xocl:irq_msix_user_setup: 0-USR-10, IRQ#140 with 0x00000000c8eac97e

Jan 6 03:42:22 ip kernel: [171304.929454] xocl:irq_msix_user_setup: 0-USR-11, IRQ#141 with 0x0000000083301800

Jan 6 03:42:22 ip kernel: [171304.929472] xocl:irq_msix_user_setup: 0-USR-12, IRQ#142 with 0x00000000b95d63c4

Jan 6 03:42:22 ip kernel: [171304.929488] xocl:irq_msix_user_setup: 0-USR-13, IRQ#143 with 0x000000005ab1eb56

Jan 6 03:42:22 ip kernel: [171304.929506] xocl:irq_msix_user_setup: 0-USR-14, IRQ#144 with 0x0000000011ac446a

Jan 6 03:42:22 ip kernel: [171304.929525] xocl:irq_msix_user_setup: 0-USR-15, IRQ#145 with 0x00000000f74a36ff

Jan 6 03:42:22 ip kernel: [171304.929561] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev dma.xdma inst 2097152

Jan 6 03:42:22 ip kernel: [171304.929563] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev dma.xdma.u inst 2097152 is active

Jan 6 03:42:22 ip kernel: [171304.929565] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev mb_scheduler.u

Jan 6 03:42:22 ip kernel: [171304.929569] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: resource [mem 0x82180000-0x82180fff]

Jan 6 03:42:22 ip kernel: [171304.929570] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: resource [mem 0x82190000-0x8219ffff]

Jan 6 03:42:22 ip kernel: [171304.929572] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: resource [irq 0-3]

Jan 6 03:42:22 ip kernel: [171304.929632] xocl_mb_sche mb_scheduler.u.5242880: dev ffff98f8f9eb0810, exec_create: CQ size is 65536

Jan 6 03:42:22 ip kernel: [171304.929643] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, exec_reset: exec_reset(2) cfg(0)

Jan 6 03:42:22 ip kernel: [171304.929645] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, exec_reset: exec_reset resets

Jan 6 03:42:22 ip kernel: [171304.929647] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, exec_reset: exec->xclbin(00000000-0000-0000-0000-000000000000),xclbin(00000000-0000-0000-0000-000000000000)

Jan 6 03:42:22 ip kernel: [171304.929753] [drm] command scheduler started

Jan 6 03:42:22 ip kernel: [171304.929767] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev mb_scheduler inst 5242880

Jan 6 03:42:22 ip kernel: [171304.929769] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev mb_scheduler.u inst 5242880 is active

Jan 6 03:42:22 ip kernel: [171304.929770] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev mailbox.u

Jan 6 03:42:22 ip kernel: [171304.930038] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_probe: successfully initialized

Jan 6 03:42:22 ip kernel: [171304.930052] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev mailbox inst 15728640

Jan 6 03:42:22 ip kernel: [171304.930139] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev mailbox.u inst 15728640 is active

Jan 6 03:42:22 ip kernel: [171304.930140] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: creating subdev icap.u

Jan 6 03:42:22 ip kernel: [171304.930172] icap.u icap.u.16777216: dev ffff98f8fa20ac10, icap_probe: successfully initialized FPGA IDCODE 0x0

Jan 6 03:42:22 ip kernel: [171304.930182] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: Created subdev icap inst 16777216

Jan 6 03:42:22 ip kernel: [171304.930183] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_create: subdev icap.u inst 16777216 is active

Jan 6 03:42:22 ip kernel: [171304.930186] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_subdev_vsec: No Vendor Specific Capability.

Jan 6 03:42:22 ip kernel: [171304.930187] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_subdev_vsec: No Vendor Specific Capability.

Jan 6 03:42:22 ip kernel: [171304.930190] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_p2p_init: Initializing P2P, bar 4, len 137438953472

Jan 6 03:42:22 ip kernel: [171304.930247] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 15 via SW

Jan 6 03:42:25 ip mpd[17775]: [0:0:1d.0] write 0 bytes out of 2104 bytes to fd 4

Jan 6 03:42:25 ip mpd[17775]: [0:0:1d.0] mpd_handleMsg thread 0 exit!!

Jan 6 03:42:25 ip mpd[17775]: 0000:00:1d.0 getMsg thread exit

Jan 6 03:42:25 ip mpd[17775]: aws mpd plugin fini called

Jan 6 03:42:25 ip mpd[17775]: ended

Jan 6 03:42:25 ip systemd[1]: Stopped Xilinx Management Proxy Daemon (MPD).

Jan 6 03:42:25 ip kernel: [171307.352907] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, timeout_msg: found outstanding msg time'd out

Jan 6 03:42:25 ip kernel: [171307.352910] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, timeout_msg: peer becomes dead

Jan 6 03:42:25 ip kernel: [171307.352964] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62

Jan 6 03:42:25 ip kernel: [171307.353438] [drm] Initialized xocl 2.5.0 20200103 for 0000:00:1d.0 on minor 0

Jan 6 03:42:25 ip kernel: [171307.353505] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 11 via SW

Jan 6 03:42:25 ip kernel: [171307.353508] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mb_connect: ch_state 0x0, ret -107

Jan 6 03:42:25 ip systemd[1]: Started Xilinx Management Proxy Daemon (MPD).

Jan 6 03:42:25 ip mpd[24133]: started

Jan 6 03:42:25 ip mpd[24133]: found mpd plugin: /opt/xilinx/xrt/lib/libmpd_plugin.so

Jan 6 03:42:25 ip mpd[24133]: aws mpd plugin init called: 0

Jan 6 03:42:25 ip mpd[24133]: create thread pair for 0000:00:1d.0

Jan 6 03:42:25 ip mpd[24133]: 1 pairs of threads running...

Jan 6 03:42:25 ip mpd[24133]: [0:0:1d.0] write 56 bytes out of 56 bytes to fd 4

Jan 6 03:42:25 ip mpd[24133]: [0:0:1d.0] msg arrived on mailbox fd 4

Jan 6 03:42:25 ip mpd[24133]: [0:0:1d.0] retrieved msg size from mailbox: 40 bytes

Jan 6 03:42:25 ip mpd[24133]: [0:0:1d.0] read 72 bytes out of 72 bytes from fd 4, valid: 1

Jan 6 03:42:25 ip mpd[24133]: [0:0:1d.0] mpd daemon: request 11 received(reqSize: 24)

Jan 6 03:42:25 ip kernel: [171307.363135] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, _xocl_drvinst_open: OPEN 1

Jan 6 03:42:25 ip kernel: [171307.363214] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, dequeue_rx_msg: peer becomes active

Jan 6 03:42:25 ip kernel: [171307.363249] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, process_request: received request from peer: 12, passed on

Jan 6 03:42:25 ip kernel: [171307.363252] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mailbox_srv: received request (12) from peer

Jan 6 03:42:25 ip kernel: [171307.363253] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mailbox_srv: mgmt driver online

Jan 6 03:42:25 ip kernel: [171307.363286] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 11 via SW

Jan 6 03:42:25 ip kernel: [171307.363384] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_read: Software TX msg is too big

Jan 6 03:42:25 ip kernel: [171307.364904] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_refresh_subdevs: get fdt from peer

Jan 6 03:42:25 ip mpd[24133]: [0:0:1d.0] msg arrived on mailbox fd 4

Jan 6 03:42:25 ip mpd[24133]: [0:0:1d.0] retrieved msg size from mailbox: 48 bytes

Jan 6 03:42:25 ip mpd[24133]: [0:0:1d.0] read 80 bytes out of 80 bytes from fd 4, valid: 1

Jan 6 03:42:25 ip kernel: [171307.365337] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 10 via SW

Jan 6 03:42:25 ip kernel: [171307.365436] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_read: Software TX msg is too big

Jan 6 03:42:49 ip mpd[24133]: [0:0:1d.0] msg arrived on mailbox fd 4

Jan 6 03:42:49 ip mpd[24133]: [0:0:1d.0] retrieved msg size from mailbox: 32 bytes

Jan 6 03:42:49 ip mpd[24133]: [0:0:1d.0] read 64 bytes out of 64 bytes from fd 4, valid: 1

Jan 6 03:42:49 ip kernel: [171331.388847] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, timeout_msg: found waiting msg time'd out

Jan 6 03:42:49 ip kernel: [171331.397224] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mb_connect: ch_state 0x0, ret -62

Jan 6 03:42:49 ip kernel: [171331.397457] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_request: sending request: 15 via SW

Jan 6 03:42:49 ip kernel: [171331.397606] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_read: Software TX msg is too big

Jan 6 03:43:13 ip kernel: [171355.428756] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, timeout_msg: found waiting msg time'd out

Jan 6 03:43:13 ip kernel: [171355.435262] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62

Jan 6 03:43:13 ip kernel: [171355.435273] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_destroy: Destroy subdev icap, cdev (null)

Jan 6 03:43:13 ip kernel: [171355.435292] icap.u icap.u.16777216: dev ffff98f8fa20ac10, icap_remove: cleaned up successfully

Jan 6 03:43:13 ip kernel: [171355.435338] xocl 0000:00:1d.0: dev ffff98f8ecf4f0a0, __xocl_subdev_destroy: Destroy subdev mailbox, cdev 00000000db05ef2b

Jan 6 03:43:13 ip kernel: [171355.435588] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, mailbox_remove: mailbox cleaned up successfully

Jan 6 03:43:13 ip kernel: [171355.435592] mailbox.u mailbox.u.15728640: dev ffff98f8f9eb5c10, xocl_drvinst_free: Wait for close 00000000726b68db

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM2C4C33DHDMICSJJWLQ4KSVZA5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEIEKQYQ#issuecomment-570992738, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBMZPZJH6UVTHI674ALLQ4KSVZANCNFSM4J26WL3A .

Here is the logs: BTW，所有的执行我这边都需要加sudo，does that matter? $ sudo systemctl stop mpd $ sudo rmmod xocl $ sudo modprobe xocl $ sudo systemctl start mpd


Jan  6 10:47:59 ip mpd[821]: mpd caught signal 15
Jan  6 10:47:59 ip systemd[1]: Stopping Xilinx Management Proxy Daemon (MPD)...
Jan  6 10:48:00 ip mpd[821]: 0000:00:1d.0 handleMsg thread exit
Jan  6 10:48:01 ip kernel: [ 3071.657631] mailbox.u mailbox.u.15728640: dev ffff8c133c8a3810, xocl_drvinst_close: CLOSE 2
Jan  6 10:48:01 ip kernel: [ 3071.657634] mailbox.u mailbox.u.15728640: dev ffff8c133c8a3810, xocl_drvinst_close: NOTIFY 0000000035587314
Jan  6 10:48:01 ip kernel: [ 3071.657724] xocl 0000:00:1d.0: dev ffff8c133e3390a0, __xocl_subdev_destroy: Destroy subdev mb_scheduler, cdev           (null)
Jan  6 10:48:01 ip kernel: [ 3071.657751] [drm] /var/lib/dkms/xrt/2.5.0/build/driver/xocl/userpf/../subdev/mb_scheduler.c:3495 scheduler thread exits with value 0
Jan  6 10:48:01 ip kernel: [ 3071.657789] [drm] command scheduler removed
Jan  6 10:48:01 ip kernel: [ 3071.657846] xocl 0000:00:1d.0: dev ffff8c133e3390a0, __xocl_subdev_destroy: Destroy subdev dma.xdma, cdev           (null)
Jan  6 10:48:01 ip mpd[821]: [0:0:1d.0] write 56 bytes out of 56 bytes to fd 4
Jan  6 10:48:01 ip mpd[821]: [0:0:1d.0] mpd_getMsg thread 0 exit!!
Jan  6 10:48:01 ip kernel: [ 3071.677818] xocl 0000:00:1d.0: dev ffff8c133e3390a0, __xocl_subdev_destroy: Destroy subdev rom, cdev           (null)
Jan  6 10:48:01 ip kernel: [ 3071.677823] rom.u rom.u.0: dev ffff8c133c8a2810, feature_rom_remove: Remove feature rom
Jan  6 10:48:04 ip kernel: [ 3074.680001] pci 0000:00:1d.0: [1d0f:1042] type 00 class 0x058000
Jan  6 10:48:04 ip kernel: [ 3074.680485] pci 0000:00:1d.0: reg 0x10: [mem 0x82000000-0x83ffffff]
Jan  6 10:48:04 ip kernel: [ 3074.680765] pci 0000:00:1d.0: reg 0x14: [mem 0x85400000-0x855fffff]
Jan  6 10:48:04 ip kernel: [ 3074.681258] pci 0000:00:1d.0: reg 0x18: [mem 0x85600000-0x8560ffff 64bit pref]
Jan  6 10:48:04 ip kernel: [ 3074.681703] pci 0000:00:1d.0: reg 0x20: [mem 0x2000000000-0x3fffffffff 64bit pref]
Jan  6 10:48:04 ip kernel: [ 3074.684582] pci 0000:00:1d.0: BAR 4: assigned [mem 0x2000000000-0x3fffffffff 64bit pref]
Jan  6 10:48:04 ip kernel: [ 3074.684872] pci 0000:00:1d.0: BAR 0: assigned [mem 0x82000000-0x83ffffff]
Jan  6 10:48:04 ip kernel: [ 3074.684954] pci 0000:00:1d.0: BAR 1: assigned [mem 0x85400000-0x855fffff]
Jan  6 10:48:04 ip kernel: [ 3074.685034] pci 0000:00:1d.0: BAR 2: assigned [mem 0x85600000-0x8560ffff 64bit pref]
Jan  6 10:48:04 ip kernel: [ 3074.685673] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.
Jan  6 10:48:04 ip kernel: [ 3074.688395] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev rom.u
Jan  6 10:48:04 ip kernel: [ 3074.688399] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x820b0000-0x820b0fff]
Jan  6 10:48:04 ip kernel: [ 3074.688455] rom.u rom.u.0: dev ffff8c132cc5d410, get_header_from_iomem: Found AWS VU9P Device without featureROM
Jan  6 10:48:04 ip kernel: [ 3074.688457] rom.u rom.u.0: dev ffff8c132cc5d410, get_header_from_iomem: Enabling AWS dynamic 5.0 Shell
Jan  6 10:48:04 ip kernel: [ 3074.688465] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: ROM magic : xlnx
Jan  6 10:48:04 ip kernel: [ 3074.688466] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: VBNV: xilinx_aws-vu9p-f1_dynamic_5_0
Jan  6 10:48:04 ip kernel: [ 3074.688468] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: DDR channel count : 4
Jan  6 10:48:04 ip kernel: [ 3074.688469] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: DDR channel size: 16 GB
Jan  6 10:48:04 ip kernel: [ 3074.688470] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: Major Version: 4
Jan  6 10:48:04 ip kernel: [ 3074.688471] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: Minor Version: 0
Jan  6 10:48:04 ip kernel: [ 3074.688472] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: IPBuildID: 43981
Jan  6 10:48:04 ip kernel: [ 3074.688474] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: TimeSinceEpoch: abcd
Jan  6 10:48:04 ip kernel: [ 3074.688475] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: FeatureBitMap: 1
Jan  6 10:48:04 ip kernel: [ 3074.688492] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev rom inst 0
Jan  6 10:48:04 ip kernel: [ 3074.688493] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev rom.u inst 0 is active
Jan  6 10:48:04 ip kernel: [ 3074.688507] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev rom.u
Jan  6 10:48:04 ip kernel: [ 3074.688508] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev dma.xdma.u
Jan  6 10:48:04 ip kernel: [ 3074.688535] xocl:xdma_device_open: xocl device 0000:00:1d.0, 0x000000005fc88d0d.
Jan  6 10:48:04 ip kernel: [ 3074.688822] xocl:map_single_bar: map bar 0, len 33554432
Jan  6 10:48:04 ip kernel: [ 3074.688823] xocl:map_single_bar: map bar 1, len 2097152
Jan  6 10:48:04 ip kernel: [ 3074.688836] xocl:map_single_bar: BAR1 at 0x85400000 mapped at 0x000000009d486877, length=2097152(/2097152)
Jan  6 10:48:04 ip kernel: [ 3074.688839] xocl:map_single_bar: map bar 2, len 65536
Jan  6 10:48:04 ip kernel: [ 3074.688851] xocl:map_single_bar: BAR2 at 0x85600000 mapped at 0x000000002b35aefd, length=65536(/65536)
Jan  6 10:48:04 ip kernel: [ 3074.688854] xocl:map_bars: config bar 2, pos 1.
Jan  6 10:48:04 ip kernel: [ 3074.688854] xocl:map_single_bar: map bar 3, len 0
Jan  6 10:48:04 ip kernel: [ 3074.688855] xocl:map_single_bar: map bar 4, len 137438953472
Jan  6 10:48:04 ip kernel: [ 3074.688855] xocl:map_single_bar: map bar 5, len 0
Jan  6 10:48:04 ip kernel: [ 3074.688856] xocl:identify_bars: xdev 0x00000000cea842f4, bars 2, config at 1.
Jan  6 10:48:04 ip kernel: [ 3074.688857] xocl:identify_bars: 2 BARs: config 1, user 1, bypass -1.
Jan  6 10:48:04 ip kernel: [ 3074.695146] xocl:enable_msi_msix: request vectors: h2c 4, c2h 4, user 16
Jan  6 10:48:04 ip kernel: [ 3074.695198] xocl:pci_keep_intx_enabled: 0000:00:1d.0: clear INTX_DISABLE, 0x406 -> 0x6.
Jan  6 10:48:04 ip kernel: [ 3074.695272] xocl:irq_msix_channel_setup: engine 0-H2C0-MM, irq#122.
Jan  6 10:48:04 ip kernel: [ 3074.695290] xocl:irq_msix_channel_setup: engine 0-H2C1-MM, irq#123.
Jan  6 10:48:04 ip kernel: [ 3074.695309] xocl:irq_msix_channel_setup: engine 0-H2C2-MM, irq#124.
Jan  6 10:48:04 ip kernel: [ 3074.695326] xocl:irq_msix_channel_setup: engine 0-H2C3-MM, irq#125.
Jan  6 10:48:04 ip kernel: [ 3074.695342] xocl:irq_msix_channel_setup: engine 0-C2H0-MM, irq#126.
Jan  6 10:48:04 ip kernel: [ 3074.695358] xocl:irq_msix_channel_setup: engine 0-C2H1-MM, irq#127.
Jan  6 10:48:04 ip kernel: [ 3074.695374] xocl:irq_msix_channel_setup: engine 0-C2H2-MM, irq#128.
Jan  6 10:48:04 ip kernel: [ 3074.695400] xocl:irq_msix_channel_setup: engine 0-C2H3-MM, irq#129.
Jan  6 10:48:04 ip kernel: [ 3074.695423] xocl:irq_msix_user_setup: 0-USR-0, IRQ#130 with 0x00000000e3335a17
Jan  6 10:48:04 ip kernel: [ 3074.695441] xocl:irq_msix_user_setup: 0-USR-1, IRQ#131 with 0x00000000c223225a
Jan  6 10:48:04 ip kernel: [ 3074.695457] xocl:irq_msix_user_setup: 0-USR-2, IRQ#132 with 0x00000000d29e436b
Jan  6 10:48:04 ip kernel: [ 3074.695473] xocl:irq_msix_user_setup: 0-USR-3, IRQ#133 with 0x000000002c7e25bb
Jan  6 10:48:04 ip kernel: [ 3074.695490] xocl:irq_msix_user_setup: 0-USR-4, IRQ#134 with 0x00000000fbcfc52a
Jan  6 10:48:04 ip kernel: [ 3074.695505] xocl:irq_msix_user_setup: 0-USR-5, IRQ#135 with 0x00000000ac42772b
Jan  6 10:48:04 ip kernel: [ 3074.695523] xocl:irq_msix_user_setup: 0-USR-6, IRQ#136 with 0x0000000005f8f660
Jan  6 10:48:04 ip kernel: [ 3074.695538] xocl:irq_msix_user_setup: 0-USR-7, IRQ#137 with 0x00000000e720e8a3
Jan  6 10:48:04 ip kernel: [ 3074.695556] xocl:irq_msix_user_setup: 0-USR-8, IRQ#138 with 0x000000007f39ab38
Jan  6 10:48:04 ip kernel: [ 3074.695572] xocl:irq_msix_user_setup: 0-USR-9, IRQ#139 with 0x00000000f462b5b6
Jan  6 10:48:04 ip kernel: [ 3074.695589] xocl:irq_msix_user_setup: 0-USR-10, IRQ#140 with 0x0000000057963f7c
Jan  6 10:48:04 ip kernel: [ 3074.695606] xocl:irq_msix_user_setup: 0-USR-11, IRQ#141 with 0x00000000bc876ace
Jan  6 10:48:04 ip kernel: [ 3074.695621] xocl:irq_msix_user_setup: 0-USR-12, IRQ#142 with 0x000000008f1a5888
Jan  6 10:48:04 ip kernel: [ 3074.695637] xocl:irq_msix_user_setup: 0-USR-13, IRQ#143 with 0x00000000b2f73562
Jan  6 10:48:04 ip kernel: [ 3074.695652] xocl:irq_msix_user_setup: 0-USR-14, IRQ#144 with 0x00000000a2937f52
Jan  6 10:48:04 ip kernel: [ 3074.695667] xocl:irq_msix_user_setup: 0-USR-15, IRQ#145 with 0x0000000043b414f7
Jan  6 10:48:04 ip kernel: [ 3074.695704] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev dma.xdma inst 2097152
Jan  6 10:48:04 ip kernel: [ 3074.695705] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev dma.xdma.u inst 2097152 is active
Jan  6 10:48:04 ip kernel: [ 3074.695707] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev mb_scheduler.u
Jan  6 10:48:04 ip kernel: [ 3074.695710] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x82180000-0x82180fff]
Jan  6 10:48:04 ip kernel: [ 3074.695711] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x82190000-0x8219ffff]
Jan  6 10:48:04 ip kernel: [ 3074.695713] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [irq 0-3]
Jan  6 10:48:04 ip kernel: [ 3074.695777] xocl_mb_sche mb_scheduler.u.5242880: dev ffff8c133e039810, exec_create: CQ size is 65536
Jan  6 10:48:04 ip kernel: [ 3074.695788] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec_reset(1) cfg(0)
Jan  6 10:48:04 ip kernel: [ 3074.695789] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec_reset resets
Jan  6 10:48:04 ip kernel: [ 3074.695791] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec->xclbin(00000000-0000-0000-0000-000000000000),xclbin(00000000-0000-0000-0000-000000000000)
Jan  6 10:48:04 ip kernel: [ 3074.695888] [drm] command scheduler started
Jan  6 10:48:04 ip kernel: [ 3074.695901] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev mb_scheduler inst 5242880
Jan  6 10:48:04 ip kernel: [ 3074.695903] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev mb_scheduler.u inst 5242880 is active
Jan  6 10:48:04 ip kernel: [ 3074.695905] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev mailbox.u
Jan  6 10:48:04 ip kernel: [ 3074.696213] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_probe: successfully initialized
Jan  6 10:48:04 ip kernel: [ 3074.696230] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev mailbox inst 15728640
Jan  6 10:48:04 ip kernel: [ 3074.696346] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev mailbox.u inst 15728640 is active
Jan  6 10:48:04 ip kernel: [ 3074.696348] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev icap.u
Jan  6 10:48:04 ip kernel: [ 3074.696384] icap.u icap.u.16777216: dev ffff8c1337aaa410, icap_probe: successfully initialized FPGA IDCODE 0x0
Jan  6 10:48:04 ip kernel: [ 3074.696394] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev icap inst 16777216
Jan  6 10:48:04 ip kernel: [ 3074.696395] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev icap.u inst 16777216 is active
Jan  6 10:48:04 ip kernel: [ 3074.696399] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.
Jan  6 10:48:04 ip kernel: [ 3074.696400] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.
Jan  6 10:48:04 ip kernel: [ 3074.696403] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_p2p_init: Initializing P2P, bar 4, len 137438953472
Jan  6 10:48:04 ip kernel: [ 3074.696461] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_request: sending request: 15 via SW
Jan  6 10:48:06 ip mpd[821]: [0:0:1d.0] write 0 bytes out of 2104 bytes to fd 4
Jan  6 10:48:06 ip mpd[821]: [0:0:1d.0] mpd_handleMsg thread 0 exit!!
Jan  6 10:48:06 ip mpd[821]: 0000:00:1d.0 getMsg thread exit
Jan  6 10:48:06 ip mpd[821]: aws mpd plugin fini called
Jan  6 10:48:06 ip mpd[821]: ended
Jan  6 10:48:06 ip systemd[1]: Stopped Xilinx Management Proxy Daemon (MPD).
Jan  6 10:48:06 ip kernel: [ 3077.117782] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, timeout_msg: found outstanding msg time'd out
Jan  6 10:48:06 ip kernel: [ 3077.117784] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, timeout_msg: peer becomes dead
Jan  6 10:48:06 ip kernel: [ 3077.117824] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62
Jan  6 10:48:06 ip kernel: [ 3077.118212] [drm] Initialized xocl 2.5.0 20200103 for 0000:00:1d.0 on minor 0
Jan  6 10:48:06 ip kernel: [ 3077.118283] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_request: sending request: 11 via SW
Jan  6 10:48:06 ip kernel: [ 3077.118286] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_connect: ch_state 0x0, ret -107
Jan  6 10:48:06 ip kernel: [ 3077.129792] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_refresh_subdevs: get fdt from peer
Jan  6 10:48:06 ip kernel: [ 3077.130232] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_request: sending request: 10 via SW
Jan  6 10:48:24 ip kernel: [ 3094.493365] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_request: sending request: 15 via SW
Jan  6 10:48:24 ip kernel: [ 3094.493368] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -107
Jan  6 10:48:24 ip kernel: [ 3094.493379] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev icap, cdev           (null)
Jan  6 10:48:24 ip kernel: [ 3094.493396] icap.u icap.u.16777216: dev ffff8c1337aaa410, icap_remove: cleaned up successfully
Jan  6 10:48:24 ip kernel: [ 3094.493446] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev mailbox, cdev 00000000266989ea
Jan  6 10:48:24 ip kernel: [ 3094.493820] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_remove: mailbox cleaned up successfully
Jan  6 10:48:24 ip kernel: [ 3094.493886] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev mb_scheduler, cdev           (null)
Jan  6 10:48:24 ip kernel: [ 3094.493922] [drm] /var/lib/dkms/xrt/2.5.0/build/driver/xocl/userpf/../subdev/mb_scheduler.c:3495 scheduler thread exits with value 0
Jan  6 10:48:24 ip kernel: [ 3094.494084] [drm] command scheduler removed
Jan  6 10:48:24 ip kernel: [ 3094.494144] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev dma.xdma, cdev           (null)
Jan  6 10:48:24 ip kernel: [ 3094.525920] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev rom, cdev           (null)
Jan  6 10:48:24 ip kernel: [ 3094.525926] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_remove: Remove feature rom
Jan  6 10:48:49 ip kernel: [ 3119.743866] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.
Jan  6 10:48:49 ip kernel: [ 3119.747047] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev rom.u
Jan  6 10:48:49 ip kernel: [ 3119.747055] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x820b0000-0x820b0fff]
Jan  6 10:48:49 ip kernel: [ 3119.747130] rom.u rom.u.0: dev ffff8c133c700410, get_header_from_iomem: Found AWS VU9P Device without featureROM
Jan  6 10:48:49 ip kernel: [ 3119.747132] rom.u rom.u.0: dev ffff8c133c700410, get_header_from_iomem: Enabling AWS dynamic 5.0 Shell
Jan  6 10:48:49 ip kernel: [ 3119.747142] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: ROM magic : xlnx
Jan  6 10:48:49 ip kernel: [ 3119.747144] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: VBNV: xilinx_aws-vu9p-f1_dynamic_5_0
Jan  6 10:48:49 ip kernel: [ 3119.747146] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: DDR channel count : 4
Jan  6 10:48:49 ip kernel: [ 3119.747148] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: DDR channel size: 16 GB
Jan  6 10:48:49 ip kernel: [ 3119.747150] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: Major Version: 4
Jan  6 10:48:49 ip kernel: [ 3119.747151] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: Minor Version: 0
Jan  6 10:48:49 ip kernel: [ 3119.747153] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: IPBuildID: 43981
Jan  6 10:48:49 ip kernel: [ 3119.747155] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: TimeSinceEpoch: abcd
Jan  6 10:48:49 ip kernel: [ 3119.747157] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: FeatureBitMap: 1
Jan  6 10:48:49 ip kernel: [ 3119.747189] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev rom inst 0
Jan  6 10:48:49 ip kernel: [ 3119.747191] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev rom.u inst 0 is active
Jan  6 10:48:49 ip kernel: [ 3119.747201] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev rom.u
Jan  6 10:48:49 ip kernel: [ 3119.747202] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev dma.xdma.u
Jan  6 10:48:49 ip kernel: [ 3119.747247] xocl:xdma_device_open: xocl device 0000:00:1d.0, 0x000000005fc88d0d.
Jan  6 10:48:49 ip kernel: [ 3119.747752] xocl:map_single_bar: map bar 0, len 33554432
Jan  6 10:48:49 ip kernel: [ 3119.747754] xocl:map_single_bar: map bar 1, len 2097152
Jan  6 10:48:49 ip kernel: [ 3119.747807] xocl:map_single_bar: BAR1 at 0x85400000 mapped at 0x00000000e79f298b, length=2097152(/2097152)
Jan  6 10:48:49 ip kernel: [ 3119.747810] xocl:map_single_bar: map bar 2, len 65536
Jan  6 10:48:49 ip kernel: [ 3119.747823] xocl:map_single_bar: BAR2 at 0x85600000 mapped at 0x00000000245e9658, length=65536(/65536)
Jan  6 10:48:49 ip kernel: [ 3119.747826] xocl:map_bars: config bar 2, pos 1.
Jan  6 10:48:49 ip kernel: [ 3119.747826] xocl:map_single_bar: map bar 3, len 0
Jan  6 10:48:49 ip kernel: [ 3119.747827] xocl:map_single_bar: map bar 4, len 137438953472
Jan  6 10:48:49 ip kernel: [ 3119.747828] xocl:map_single_bar: map bar 5, len 0
Jan  6 10:48:49 ip kernel: [ 3119.747829] xocl:identify_bars: xdev 0x000000008517ec9d, bars 2, config at 1.
Jan  6 10:48:49 ip kernel: [ 3119.747829] xocl:identify_bars: 2 BARs: config 1, user 1, bypass -1.
Jan  6 10:48:49 ip kernel: [ 3119.755625] xocl:enable_msi_msix: request vectors: h2c 4, c2h 4, user 16
Jan  6 10:48:49 ip kernel: [ 3119.755695] xocl:pci_keep_intx_enabled: 0000:00:1d.0: clear INTX_DISABLE, 0x406 -> 0x6.
Jan  6 10:48:49 ip kernel: [ 3119.755785] xocl:irq_msix_channel_setup: engine 0-H2C0-MM, irq#122.
Jan  6 10:48:49 ip kernel: [ 3119.755804] xocl:irq_msix_channel_setup: engine 0-H2C1-MM, irq#123.
Jan  6 10:48:49 ip kernel: [ 3119.755820] xocl:irq_msix_channel_setup: engine 0-H2C2-MM, irq#124.
Jan  6 10:48:49 ip kernel: [ 3119.755842] xocl:irq_msix_channel_setup: engine 0-H2C3-MM, irq#125.
Jan  6 10:48:49 ip kernel: [ 3119.755858] xocl:irq_msix_channel_setup: engine 0-C2H0-MM, irq#126.
Jan  6 10:48:49 ip kernel: [ 3119.755879] xocl:irq_msix_channel_setup: engine 0-C2H1-MM, irq#127.
Jan  6 10:48:49 ip kernel: [ 3119.755896] xocl:irq_msix_channel_setup: engine 0-C2H2-MM, irq#128.
Jan  6 10:48:49 ip kernel: [ 3119.755916] xocl:irq_msix_channel_setup: engine 0-C2H3-MM, irq#129.
Jan  6 10:48:49 ip kernel: [ 3119.755932] xocl:irq_msix_user_setup: 0-USR-0, IRQ#130 with 0x0000000083739a6d
Jan  6 10:48:49 ip kernel: [ 3119.755949] xocl:irq_msix_user_setup: 0-USR-1, IRQ#131 with 0x000000008ae17967
Jan  6 10:48:49 ip kernel: [ 3119.755970] xocl:irq_msix_user_setup: 0-USR-2, IRQ#132 with 0x000000005e54f2c5
Jan  6 10:48:49 ip kernel: [ 3119.755986] xocl:irq_msix_user_setup: 0-USR-3, IRQ#133 with 0x000000000ee8c643
Jan  6 10:48:49 ip kernel: [ 3119.756006] xocl:irq_msix_user_setup: 0-USR-4, IRQ#134 with 0x0000000039ae046f
Jan  6 10:48:49 ip kernel: [ 3119.756021] xocl:irq_msix_user_setup: 0-USR-5, IRQ#135 with 0x00000000b27b8661
Jan  6 10:48:49 ip kernel: [ 3119.756045] xocl:irq_msix_user_setup: 0-USR-6, IRQ#136 with 0x00000000d213e848
Jan  6 10:48:49 ip kernel: [ 3119.756060] xocl:irq_msix_user_setup: 0-USR-7, IRQ#137 with 0x00000000287f528b
Jan  6 10:48:49 ip kernel: [ 3119.756075] xocl:irq_msix_user_setup: 0-USR-8, IRQ#138 with 0x000000002ca52cbc
Jan  6 10:48:49 ip kernel: [ 3119.756095] xocl:irq_msix_user_setup: 0-USR-9, IRQ#139 with 0x0000000008836bf9
Jan  6 10:48:49 ip kernel: [ 3119.756110] xocl:irq_msix_user_setup: 0-USR-10, IRQ#140 with 0x000000001a1c86d2
Jan  6 10:48:49 ip kernel: [ 3119.756131] xocl:irq_msix_user_setup: 0-USR-11, IRQ#141 with 0x00000000daff9f21
Jan  6 10:48:49 ip kernel: [ 3119.756147] xocl:irq_msix_user_setup: 0-USR-12, IRQ#142 with 0x000000005b6f162f
Jan  6 10:48:49 ip kernel: [ 3119.756167] xocl:irq_msix_user_setup: 0-USR-13, IRQ#143 with 0x00000000a99095e8
Jan  6 10:48:49 ip kernel: [ 3119.756182] xocl:irq_msix_user_setup: 0-USR-14, IRQ#144 with 0x00000000a2557c42
Jan  6 10:48:49 ip kernel: [ 3119.756197] xocl:irq_msix_user_setup: 0-USR-15, IRQ#145 with 0x0000000050414470
Jan  6 10:48:49 ip kernel: [ 3119.756237] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev dma.xdma inst 2097152
Jan  6 10:48:49 ip kernel: [ 3119.756240] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev dma.xdma.u inst 2097152 is active
Jan  6 10:48:49 ip kernel: [ 3119.756242] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev mb_scheduler.u
Jan  6 10:48:49 ip kernel: [ 3119.756248] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x82180000-0x82180fff]
Jan  6 10:48:49 ip kernel: [ 3119.756249] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x82190000-0x8219ffff]
Jan  6 10:48:49 ip kernel: [ 3119.756252] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [irq 0-3]
Jan  6 10:48:49 ip kernel: [ 3119.756335] xocl_mb_sche mb_scheduler.u.5242880: dev ffff8c133c705010, exec_create: CQ size is 65536
Jan  6 10:48:49 ip kernel: [ 3119.756347] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec_reset(0) cfg(0)
Jan  6 10:48:49 ip kernel: [ 3119.756349] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec_reset resets
Jan  6 10:48:49 ip kernel: [ 3119.756352] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec->xclbin(00000000-0000-0000-0000-000000000000),xclbin(00000000-0000-0000-0000-000000000000)
Jan  6 10:48:49 ip kernel: [ 3119.756433] [drm] command scheduler started
Jan  6 10:48:49 ip kernel: [ 3119.756473] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev mb_scheduler inst 5242880
Jan  6 10:48:49 ip kernel: [ 3119.756476] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev mb_scheduler.u inst 5242880 is active
Jan  6 10:48:49 ip kernel: [ 3119.756478] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev mailbox.u
Jan  6 10:48:49 ip kernel: [ 3119.756795] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_probe: successfully initialized
Jan  6 10:48:49 ip kernel: [ 3119.756810] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev mailbox inst 15728640
Jan  6 10:48:49 ip kernel: [ 3119.756917] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev mailbox.u inst 15728640 is active
Jan  6 10:48:49 ip kernel: [ 3119.756919] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev icap.u
Jan  6 10:48:49 ip kernel: [ 3119.756956] icap.u icap.u.16777216: dev ffff8c133e189410, icap_probe: successfully initialized FPGA IDCODE 0x0
Jan  6 10:48:49 ip kernel: [ 3119.756963] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev icap inst 16777216
Jan  6 10:48:49 ip kernel: [ 3119.756965] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev icap.u inst 16777216 is active
Jan  6 10:48:49 ip kernel: [ 3119.756969] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.
Jan  6 10:48:49 ip kernel: [ 3119.756970] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.
Jan  6 10:48:49 ip kernel: [ 3119.756972] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_p2p_init: Initializing P2P, bar 4, len 137438953472
Jan  6 10:48:49 ip kernel: [ 3119.757038] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 15 via SW
Jan  6 10:48:51 ip kernel: [ 3122.181787] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, timeout_msg: found outstanding msg time'd out
Jan  6 10:48:51 ip kernel: [ 3122.181789] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, timeout_msg: peer becomes dead
Jan  6 10:48:51 ip kernel: [ 3122.181823] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62
Jan  6 10:48:51 ip kernel: [ 3122.182191] [drm] Initialized xocl 2.5.0 20200103 for 0000:00:1d.0 on minor 0
Jan  6 10:48:51 ip kernel: [ 3122.182260] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 11 via SW
Jan  6 10:48:51 ip kernel: [ 3122.182263] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_connect: ch_state 0x0, ret -107
Jan  6 10:48:51 ip kernel: [ 3122.189798] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_refresh_subdevs: get fdt from peer
Jan  6 10:48:51 ip kernel: [ 3122.190203] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 10 via SW
Jan  6 10:49:22 ip systemd[1]: Started Xilinx Management Proxy Daemon (MPD).
Jan  6 10:49:22 ip mpd[1859]: started
Jan  6 10:49:22 ip mpd[1859]: found mpd plugin: /opt/xilinx/xrt/lib/libmpd_plugin.so
Jan  6 10:49:22 ip mpd[1859]: aws mpd plugin init called: 0
Jan  6 10:49:22 ip mpd[1859]: create thread pair for 0000:00:1d.0
Jan  6 10:49:22 ip mpd[1859]: 1 pairs of threads running...
Jan  6 10:49:22 ip mpd[1859]: [0:0:1d.0] write 56 bytes out of 56 bytes to fd 4
Jan  6 10:49:22 ip mpd[1859]: [0:0:1d.0] msg arrived on mailbox fd 4
Jan  6 10:49:22 ip mpd[1859]: [0:0:1d.0] retrieved msg size from mailbox: 40 bytes
Jan  6 10:49:22 ip mpd[1859]: [0:0:1d.0] read 72 bytes out of 72 bytes from fd 4, valid: 1
Jan  6 10:49:22 ip mpd[1859]: [0:0:1d.0] mpd daemon: request 11 received(reqSize: 24)
Jan  6 10:49:22 ip kernel: [ 3152.625936] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, _xocl_drvinst_open: OPEN 1
Jan  6 10:49:22 ip kernel: [ 3152.625999] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, dequeue_rx_msg: peer becomes active
Jan  6 10:49:22 ip kernel: [ 3152.626035] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, process_request: received request from peer: 12, passed on
Jan  6 10:49:22 ip kernel: [ 3152.626038] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mailbox_srv: received request (12) from peer
Jan  6 10:49:22 ip kernel: [ 3152.626040] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mailbox_srv: mgmt driver online
Jan  6 10:49:22 ip kernel: [ 3152.626103] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 11 via SW
Jan  6 10:49:22 ip kernel: [ 3152.626235] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_read: Software TX msg is too big
Jan  6 10:49:25 ip mpd[1859]: [0:0:1d.0] msg arrived on mailbox fd 4
Jan  6 10:49:25 ip mpd[1859]: [0:0:1d.0] retrieved msg size from mailbox: 32 bytes
Jan  6 10:49:25 ip mpd[1859]: [0:0:1d.0] read 64 bytes out of 64 bytes from fd 4, valid: 1
Jan  6 10:49:25 ip kernel: [ 3155.623617] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 15 via SW
Jan  6 10:49:25 ip kernel: [ 3155.623742] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_read: Software TX msg is too big
Jan  6 10:49:46 ip kernel: [ 3176.649814] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, timeout_msg: found waiting msg time'd out
Jan  6 10:49:46 ip kernel: [ 3176.659006] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_connect: ch_state 0x0, ret -62
Jan  6 10:49:49 ip kernel: [ 3179.625808] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, timeout_msg: found waiting msg time'd out
Jan  6 10:49:49 ip kernel: [ 3179.635898] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62
Jan  6 10:49:49 ip kernel: [ 3179.635912] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev icap, cdev           (null)
Jan  6 10:49:49 ip kernel: [ 3179.635935] icap.u icap.u.16777216: dev ffff8c133e189410, icap_remove: cleaned up successfully
Jan  6 10:49:49 ip kernel: [ 3179.635990] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev mailbox, cdev 0000000086a62c1b
Jan  6 10:49:49 ip kernel: [ 3179.636308] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_remove: mailbox cleaned up successfully
Jan  6 10:49:49 ip kernel: [ 3179.636313] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, xocl_drvinst_free: Wait for close 00000000067277db

我在一个F1 上试了一下，xrt的master最新和2019.2分支都是好的，正常的log大致是这样的。

systemctl stop mpd

Jan 6 21:30:28 ip-172-31-70-242 systemd[1]: Stopped Xilinx Management Proxy Daemon (MPD).

rmmod xocl

... Jan 6 21:31:21 ip-172-31-70-242 kernel: [4649834.709001] [drm] /var/lib/dkms/xrt/2.5.0/build/driver/xocl/userpf/../subdev/mb_scheduler.c:3495 scheduler thread exits with value 0 Jan 6 21:31:21 ip-172-31-70-242 kernel: [4649834.709052] [drm] command scheduler removed Jan 6 21:31:21 ip-172-31-70-242 kernel: [4649834.709088] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_subdev_destroy: Destroy subdev dma.xdma, cdev (null) Jan 6 21:31:21 ip-172-31-70-242 kernel: [4649834.724580] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_subdev_destroy: Destroy subdev rom, cdev (null) Jan 6 21:31:21 ip-172-31-70-242 kernel: [4649834.724594] rom.u rom.u.0: dev ffff881df49fa810, feature_rom_remove: Remove feature rom

modprobe xocl

Jan 6 21:33:03 ip-172-31-70-242 kernel: [4649936.423315] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_subdev_vsec: No Vendor Specific Capability. Jan 6 21:33:03 ip-172-31-70-242 kernel: [4649936.425170] xocl 0000:00:1d.0: dev ffff881df8167098, __xocl_subdev_create: creating subdev rom.u ... Jan 6 21:33:03 ip-172-31-70-242 kernel: [4649936.435390] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_p2p_init: Initializing P2P, bar 4, len 137438953472 Jan 6 21:33:03 ip-172-31-70-242 kernel: [4649936.435426] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, mailbox_request: sending request: 15 via SW Jan 6 21:33:05 ip-172-31-70-242 kernel: [4649938.452622] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, timeout_msg: found outstanding msg time'd out Jan 6 21:33:05 ip-172-31-70-242 kernel: [4649938.452630] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, timeout_msg: peer becomes dead Jan 6 21:33:05 ip-172-31-70-242 kernel: [4649938.452684] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62 Jan 6 21:33:05 ip-172-31-70-242 kernel: [4649938.453320] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, mailbox_request: sending request: 11 via SW Jan 6 21:33:05 ip-172-31-70-242 kernel: [4649938.453324] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_mb_connect: ch_state 0x0, ret -107 Jan 6 21:33:05 ip-172-31-70-242 kernel: [4649938.456616] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_refresh_subdevs: get fdt from peer Jan 6 21:33:05 ip-172-31-70-242 kernel: [4649938.456925] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, mailbox_request: sending request: 10 via SW

systemctl start mpd

Jan 6 21:34:35 ip-172-31-70-242 systemd[1]: Started Xilinx Management Proxy Daemon (MPD). Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: started Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: found mpd plugin: /opt/xilinx/xrt/lib/libmpd_plugin.so Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: aws mpd plugin init called: 0 Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: create thread pair for 0000:00:1d.0 Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: 1 pairs of threads running... Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] write 56 bytes out of 56 bytes to fd 4 Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] msg arrived on mailbox fd 4 Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.287460] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, _xocl_drvinst_open: OPEN 1 Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.287499] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, dequeue_rx_msg: peer becomes active Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.287532] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, process_request: received request from peer: 12, passed on Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.287537] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_mailbox_srv: received request (12) from peer Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.287539] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_mailbox_srv: mgmt driver online Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.287938] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, mailbox_request: sending request: 11 via SW Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.288024] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, mailbox_read: Software TX msg is too big Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] retrieved msg size from mailbox: 40 bytes Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] read 72 bytes out of 72 bytes from fd 4, valid: 1 Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] mpd daemon: request 11 received(reqSize: 24) Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] write 2104 bytes out of 2104 bytes to fd 4 Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.288790] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_mb_connect: ch_state 0x1, ret 0 Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.292686] xocl 0000:00:1d.0: dev ffff881df8167098, xocl_refresh_subdevs: get fdt from peer Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] msg arrived on mailbox fd 4 Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] retrieved msg size from mailbox: 48 bytes Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] read 80 bytes out of 80 bytes from fd 4, valid: 1 Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] mpd daemon: request 10 received(reqSize: 32) Jan 6 21:34:35 ip-172-31-70-242 mpd[6273]: [0:0:1d.0] write 524360 bytes out of 524360 bytes to fd 4 Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.292989] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, mailbox_request: sending request: 10 via SW Jan 6 21:34:35 ip-172-31-70-242 kernel: [4650028.293064] mailbox.u mailbox.u.15728640: dev ffff881cc883a010, mailbox_read: Software TX msg is too big

你的log，尤其在停掉mpd和卸载xocl这两步，我看不太明白。请确认，

停掉mpd后，检查‘ps -ef|grep mpd', 没有mpd进程存在了
卸载xocl以后，检查’ lsmod |grep xocl'，没有xocl module了
你把每一步之后的log和我的比较一下，看看有什么问题
你把你编译和安装xrt的history贴一下，我看看有什么问题或者有什么办法可以让我登录上去看一下

On Mon, Jan 6, 2020 at 2:57 AM Liang Ma notifications@github.com wrote:

Here is the logs: BTW，所有的执行我这边都需要加sudo，does that matter? $ sudo systemctl stop mpd $ sudo rmmod xocl $ sudo modprobe xocl $ sudo systemctl start mpd

Jan 6 10:47:59 ip mpd[821]: mpd caught signal 15

Jan 6 10:47:59 ip systemd[1]: Stopping Xilinx Management Proxy Daemon (MPD)...

Jan 6 10:48:00 ip mpd[821]: 0000:00:1d.0 handleMsg thread exit

Jan 6 10:48:01 ip kernel: [ 3071.657631] mailbox.u mailbox.u.15728640: dev ffff8c133c8a3810, xocl_drvinst_close: CLOSE 2

Jan 6 10:48:01 ip kernel: [ 3071.657634] mailbox.u mailbox.u.15728640: dev ffff8c133c8a3810, xocl_drvinst_close: NOTIFY 0000000035587314

Jan 6 10:48:01 ip kernel: [ 3071.657724] xocl 0000:00:1d.0: dev ffff8c133e3390a0, __xocl_subdev_destroy: Destroy subdev mb_scheduler, cdev (null)

Jan 6 10:48:01 ip kernel: [ 3071.657751] [drm] /var/lib/dkms/xrt/2.5.0/build/driver/xocl/userpf/../subdev/mb_scheduler.c:3495 scheduler thread exits with value 0

Jan 6 10:48:01 ip kernel: [ 3071.657789] [drm] command scheduler removed

Jan 6 10:48:01 ip kernel: [ 3071.657846] xocl 0000:00:1d.0: dev ffff8c133e3390a0, __xocl_subdev_destroy: Destroy subdev dma.xdma, cdev (null)

Jan 6 10:48:01 ip mpd[821]: [0:0:1d.0] write 56 bytes out of 56 bytes to fd 4

Jan 6 10:48:01 ip mpd[821]: [0:0:1d.0] mpd_getMsg thread 0 exit!!

Jan 6 10:48:01 ip kernel: [ 3071.677818] xocl 0000:00:1d.0: dev ffff8c133e3390a0, __xocl_subdev_destroy: Destroy subdev rom, cdev (null)

Jan 6 10:48:01 ip kernel: [ 3071.677823] rom.u rom.u.0: dev ffff8c133c8a2810, feature_rom_remove: Remove feature rom

Jan 6 10:48:04 ip kernel: [ 3074.680001] pci 0000:00:1d.0: [1d0f:1042] type 00 class 0x058000

Jan 6 10:48:04 ip kernel: [ 3074.680485] pci 0000:00:1d.0: reg 0x10: [mem 0x82000000-0x83ffffff]

Jan 6 10:48:04 ip kernel: [ 3074.680765] pci 0000:00:1d.0: reg 0x14: [mem 0x85400000-0x855fffff]

Jan 6 10:48:04 ip kernel: [ 3074.681258] pci 0000:00:1d.0: reg 0x18: [mem 0x85600000-0x8560ffff 64bit pref]

Jan 6 10:48:04 ip kernel: [ 3074.681703] pci 0000:00:1d.0: reg 0x20: [mem 0x2000000000-0x3fffffffff 64bit pref]

Jan 6 10:48:04 ip kernel: [ 3074.684582] pci 0000:00:1d.0: BAR 4: assigned [mem 0x2000000000-0x3fffffffff 64bit pref]

Jan 6 10:48:04 ip kernel: [ 3074.684872] pci 0000:00:1d.0: BAR 0: assigned [mem 0x82000000-0x83ffffff]

Jan 6 10:48:04 ip kernel: [ 3074.684954] pci 0000:00:1d.0: BAR 1: assigned [mem 0x85400000-0x855fffff]

Jan 6 10:48:04 ip kernel: [ 3074.685034] pci 0000:00:1d.0: BAR 2: assigned [mem 0x85600000-0x8560ffff 64bit pref]

Jan 6 10:48:04 ip kernel: [ 3074.685673] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.

Jan 6 10:48:04 ip kernel: [ 3074.688395] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev rom.u

Jan 6 10:48:04 ip kernel: [ 3074.688399] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x820b0000-0x820b0fff]

Jan 6 10:48:04 ip kernel: [ 3074.688455] rom.u rom.u.0: dev ffff8c132cc5d410, get_header_from_iomem: Found AWS VU9P Device without featureROM

Jan 6 10:48:04 ip kernel: [ 3074.688457] rom.u rom.u.0: dev ffff8c132cc5d410, get_header_from_iomem: Enabling AWS dynamic 5.0 Shell

Jan 6 10:48:04 ip kernel: [ 3074.688465] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: ROM magic : xlnx

Jan 6 10:48:04 ip kernel: [ 3074.688466] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: VBNV: xilinx_aws-vu9p-f1_dynamic_5_0

Jan 6 10:48:04 ip kernel: [ 3074.688468] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: DDR channel count : 4

Jan 6 10:48:04 ip kernel: [ 3074.688469] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: DDR channel size: 16 GB

Jan 6 10:48:04 ip kernel: [ 3074.688470] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: Major Version: 4

Jan 6 10:48:04 ip kernel: [ 3074.688471] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: Minor Version: 0

Jan 6 10:48:04 ip kernel: [ 3074.688472] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: IPBuildID: 43981

Jan 6 10:48:04 ip kernel: [ 3074.688474] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: TimeSinceEpoch: abcd

Jan 6 10:48:04 ip kernel: [ 3074.688475] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_probe: FeatureBitMap: 1

Jan 6 10:48:04 ip kernel: [ 3074.688492] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev rom inst 0

Jan 6 10:48:04 ip kernel: [ 3074.688493] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev rom.u inst 0 is active

Jan 6 10:48:04 ip kernel: [ 3074.688507] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev rom.u

Jan 6 10:48:04 ip kernel: [ 3074.688508] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev dma.xdma.u

Jan 6 10:48:04 ip kernel: [ 3074.688535] xocl:xdma_device_open: xocl device 0000:00:1d.0, 0x000000005fc88d0d.

Jan 6 10:48:04 ip kernel: [ 3074.688822] xocl:map_single_bar: map bar 0, len 33554432

Jan 6 10:48:04 ip kernel: [ 3074.688823] xocl:map_single_bar: map bar 1, len 2097152

Jan 6 10:48:04 ip kernel: [ 3074.688836] xocl:map_single_bar: BAR1 at 0x85400000 mapped at 0x000000009d486877, length=2097152(/2097152)

Jan 6 10:48:04 ip kernel: [ 3074.688839] xocl:map_single_bar: map bar 2, len 65536

Jan 6 10:48:04 ip kernel: [ 3074.688851] xocl:map_single_bar: BAR2 at 0x85600000 mapped at 0x000000002b35aefd, length=65536(/65536)

Jan 6 10:48:04 ip kernel: [ 3074.688854] xocl:map_bars: config bar 2, pos 1.

Jan 6 10:48:04 ip kernel: [ 3074.688854] xocl:map_single_bar: map bar 3, len 0

Jan 6 10:48:04 ip kernel: [ 3074.688855] xocl:map_single_bar: map bar 4, len 137438953472

Jan 6 10:48:04 ip kernel: [ 3074.688855] xocl:map_single_bar: map bar 5, len 0

Jan 6 10:48:04 ip kernel: [ 3074.688856] xocl:identify_bars: xdev 0x00000000cea842f4, bars 2, config at 1.

Jan 6 10:48:04 ip kernel: [ 3074.688857] xocl:identify_bars: 2 BARs: config 1, user 1, bypass -1.

Jan 6 10:48:04 ip kernel: [ 3074.695146] xocl:enable_msi_msix: request vectors: h2c 4, c2h 4, user 16

Jan 6 10:48:04 ip kernel: [ 3074.695198] xocl:pci_keep_intx_enabled: 0000:00:1d.0: clear INTX_DISABLE, 0x406 -> 0x6.

Jan 6 10:48:04 ip kernel: [ 3074.695272] xocl:irq_msix_channel_setup: engine 0-H2C0-MM, irq#122.

Jan 6 10:48:04 ip kernel: [ 3074.695290] xocl:irq_msix_channel_setup: engine 0-H2C1-MM, irq#123.

Jan 6 10:48:04 ip kernel: [ 3074.695309] xocl:irq_msix_channel_setup: engine 0-H2C2-MM, irq#124.

Jan 6 10:48:04 ip kernel: [ 3074.695326] xocl:irq_msix_channel_setup: engine 0-H2C3-MM, irq#125.

Jan 6 10:48:04 ip kernel: [ 3074.695342] xocl:irq_msix_channel_setup: engine 0-C2H0-MM, irq#126.

Jan 6 10:48:04 ip kernel: [ 3074.695358] xocl:irq_msix_channel_setup: engine 0-C2H1-MM, irq#127.

Jan 6 10:48:04 ip kernel: [ 3074.695374] xocl:irq_msix_channel_setup: engine 0-C2H2-MM, irq#128.

Jan 6 10:48:04 ip kernel: [ 3074.695400] xocl:irq_msix_channel_setup: engine 0-C2H3-MM, irq#129.

Jan 6 10:48:04 ip kernel: [ 3074.695423] xocl:irq_msix_user_setup: 0-USR-0, IRQ#130 with 0x00000000e3335a17

Jan 6 10:48:04 ip kernel: [ 3074.695441] xocl:irq_msix_user_setup: 0-USR-1, IRQ#131 with 0x00000000c223225a

Jan 6 10:48:04 ip kernel: [ 3074.695457] xocl:irq_msix_user_setup: 0-USR-2, IRQ#132 with 0x00000000d29e436b

Jan 6 10:48:04 ip kernel: [ 3074.695473] xocl:irq_msix_user_setup: 0-USR-3, IRQ#133 with 0x000000002c7e25bb

Jan 6 10:48:04 ip kernel: [ 3074.695490] xocl:irq_msix_user_setup: 0-USR-4, IRQ#134 with 0x00000000fbcfc52a

Jan 6 10:48:04 ip kernel: [ 3074.695505] xocl:irq_msix_user_setup: 0-USR-5, IRQ#135 with 0x00000000ac42772b

Jan 6 10:48:04 ip kernel: [ 3074.695523] xocl:irq_msix_user_setup: 0-USR-6, IRQ#136 with 0x0000000005f8f660

Jan 6 10:48:04 ip kernel: [ 3074.695538] xocl:irq_msix_user_setup: 0-USR-7, IRQ#137 with 0x00000000e720e8a3

Jan 6 10:48:04 ip kernel: [ 3074.695556] xocl:irq_msix_user_setup: 0-USR-8, IRQ#138 with 0x000000007f39ab38

Jan 6 10:48:04 ip kernel: [ 3074.695572] xocl:irq_msix_user_setup: 0-USR-9, IRQ#139 with 0x00000000f462b5b6

Jan 6 10:48:04 ip kernel: [ 3074.695589] xocl:irq_msix_user_setup: 0-USR-10, IRQ#140 with 0x0000000057963f7c

Jan 6 10:48:04 ip kernel: [ 3074.695606] xocl:irq_msix_user_setup: 0-USR-11, IRQ#141 with 0x00000000bc876ace

Jan 6 10:48:04 ip kernel: [ 3074.695621] xocl:irq_msix_user_setup: 0-USR-12, IRQ#142 with 0x000000008f1a5888

Jan 6 10:48:04 ip kernel: [ 3074.695637] xocl:irq_msix_user_setup: 0-USR-13, IRQ#143 with 0x00000000b2f73562

Jan 6 10:48:04 ip kernel: [ 3074.695652] xocl:irq_msix_user_setup: 0-USR-14, IRQ#144 with 0x00000000a2937f52

Jan 6 10:48:04 ip kernel: [ 3074.695667] xocl:irq_msix_user_setup: 0-USR-15, IRQ#145 with 0x0000000043b414f7

Jan 6 10:48:04 ip kernel: [ 3074.695704] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev dma.xdma inst 2097152

Jan 6 10:48:04 ip kernel: [ 3074.695705] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev dma.xdma.u inst 2097152 is active

Jan 6 10:48:04 ip kernel: [ 3074.695707] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev mb_scheduler.u

Jan 6 10:48:04 ip kernel: [ 3074.695710] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x82180000-0x82180fff]

Jan 6 10:48:04 ip kernel: [ 3074.695711] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x82190000-0x8219ffff]

Jan 6 10:48:04 ip kernel: [ 3074.695713] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [irq 0-3]

Jan 6 10:48:04 ip kernel: [ 3074.695777] xocl_mb_sche mb_scheduler.u.5242880: dev ffff8c133e039810, exec_create: CQ size is 65536

Jan 6 10:48:04 ip kernel: [ 3074.695788] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec_reset(1) cfg(0)

Jan 6 10:48:04 ip kernel: [ 3074.695789] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec_reset resets

Jan 6 10:48:04 ip kernel: [ 3074.695791] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec->xclbin(00000000-0000-0000-0000-000000000000),xclbin(00000000-0000-0000-0000-000000000000)

Jan 6 10:48:04 ip kernel: [ 3074.695888] [drm] command scheduler started

Jan 6 10:48:04 ip kernel: [ 3074.695901] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev mb_scheduler inst 5242880

Jan 6 10:48:04 ip kernel: [ 3074.695903] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev mb_scheduler.u inst 5242880 is active

Jan 6 10:48:04 ip kernel: [ 3074.695905] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev mailbox.u

Jan 6 10:48:04 ip kernel: [ 3074.696213] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_probe: successfully initialized

Jan 6 10:48:04 ip kernel: [ 3074.696230] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev mailbox inst 15728640

Jan 6 10:48:04 ip kernel: [ 3074.696346] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev mailbox.u inst 15728640 is active

Jan 6 10:48:04 ip kernel: [ 3074.696348] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev icap.u

Jan 6 10:48:04 ip kernel: [ 3074.696384] icap.u icap.u.16777216: dev ffff8c1337aaa410, icap_probe: successfully initialized FPGA IDCODE 0x0

Jan 6 10:48:04 ip kernel: [ 3074.696394] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev icap inst 16777216

Jan 6 10:48:04 ip kernel: [ 3074.696395] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev icap.u inst 16777216 is active

Jan 6 10:48:04 ip kernel: [ 3074.696399] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.

Jan 6 10:48:04 ip kernel: [ 3074.696400] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.

Jan 6 10:48:04 ip kernel: [ 3074.696403] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_p2p_init: Initializing P2P, bar 4, len 137438953472

Jan 6 10:48:04 ip kernel: [ 3074.696461] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_request: sending request: 15 via SW

Jan 6 10:48:06 ip mpd[821]: [0:0:1d.0] write 0 bytes out of 2104 bytes to fd 4

Jan 6 10:48:06 ip mpd[821]: [0:0:1d.0] mpd_handleMsg thread 0 exit!!

Jan 6 10:48:06 ip mpd[821]: 0000:00:1d.0 getMsg thread exit

Jan 6 10:48:06 ip mpd[821]: aws mpd plugin fini called

Jan 6 10:48:06 ip mpd[821]: ended

Jan 6 10:48:06 ip systemd[1]: Stopped Xilinx Management Proxy Daemon (MPD).

Jan 6 10:48:06 ip kernel: [ 3077.117782] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, timeout_msg: found outstanding msg time'd out

Jan 6 10:48:06 ip kernel: [ 3077.117784] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, timeout_msg: peer becomes dead

Jan 6 10:48:06 ip kernel: [ 3077.117824] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62

Jan 6 10:48:06 ip kernel: [ 3077.118212] [drm] Initialized xocl 2.5.0 20200103 for 0000:00:1d.0 on minor 0

Jan 6 10:48:06 ip kernel: [ 3077.118283] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_request: sending request: 11 via SW

Jan 6 10:48:06 ip kernel: [ 3077.118286] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_connect: ch_state 0x0, ret -107

Jan 6 10:48:06 ip kernel: [ 3077.129792] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_refresh_subdevs: get fdt from peer

Jan 6 10:48:06 ip kernel: [ 3077.130232] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_request: sending request: 10 via SW

Jan 6 10:48:24 ip kernel: [ 3094.493365] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_request: sending request: 15 via SW

Jan 6 10:48:24 ip kernel: [ 3094.493368] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -107

Jan 6 10:48:24 ip kernel: [ 3094.493379] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev icap, cdev (null)

Jan 6 10:48:24 ip kernel: [ 3094.493396] icap.u icap.u.16777216: dev ffff8c1337aaa410, icap_remove: cleaned up successfully

Jan 6 10:48:24 ip kernel: [ 3094.493446] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev mailbox, cdev 00000000266989ea

Jan 6 10:48:24 ip kernel: [ 3094.493820] mailbox.u mailbox.u.15728640: dev ffff8c133e03dc10, mailbox_remove: mailbox cleaned up successfully

Jan 6 10:48:24 ip kernel: [ 3094.493886] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev mb_scheduler, cdev (null)

Jan 6 10:48:24 ip kernel: [ 3094.493922] [drm] /var/lib/dkms/xrt/2.5.0/build/driver/xocl/userpf/../subdev/mb_scheduler.c:3495 scheduler thread exits with value 0

Jan 6 10:48:24 ip kernel: [ 3094.494084] [drm] command scheduler removed

Jan 6 10:48:24 ip kernel: [ 3094.494144] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev dma.xdma, cdev (null)

Jan 6 10:48:24 ip kernel: [ 3094.525920] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev rom, cdev (null)

Jan 6 10:48:24 ip kernel: [ 3094.525926] rom.u rom.u.0: dev ffff8c132cc5d410, feature_rom_remove: Remove feature rom

Jan 6 10:48:49 ip kernel: [ 3119.743866] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.

Jan 6 10:48:49 ip kernel: [ 3119.747047] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev rom.u

Jan 6 10:48:49 ip kernel: [ 3119.747055] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x820b0000-0x820b0fff]

Jan 6 10:48:49 ip kernel: [ 3119.747130] rom.u rom.u.0: dev ffff8c133c700410, get_header_from_iomem: Found AWS VU9P Device without featureROM

Jan 6 10:48:49 ip kernel: [ 3119.747132] rom.u rom.u.0: dev ffff8c133c700410, get_header_from_iomem: Enabling AWS dynamic 5.0 Shell

Jan 6 10:48:49 ip kernel: [ 3119.747142] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: ROM magic : xlnx

Jan 6 10:48:49 ip kernel: [ 3119.747144] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: VBNV: xilinx_aws-vu9p-f1_dynamic_5_0

Jan 6 10:48:49 ip kernel: [ 3119.747146] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: DDR channel count : 4

Jan 6 10:48:49 ip kernel: [ 3119.747148] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: DDR channel size: 16 GB

Jan 6 10:48:49 ip kernel: [ 3119.747150] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: Major Version: 4

Jan 6 10:48:49 ip kernel: [ 3119.747151] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: Minor Version: 0

Jan 6 10:48:49 ip kernel: [ 3119.747153] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: IPBuildID: 43981

Jan 6 10:48:49 ip kernel: [ 3119.747155] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: TimeSinceEpoch: abcd

Jan 6 10:48:49 ip kernel: [ 3119.747157] rom.u rom.u.0: dev ffff8c133c700410, feature_rom_probe: FeatureBitMap: 1

Jan 6 10:48:49 ip kernel: [ 3119.747189] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev rom inst 0

Jan 6 10:48:49 ip kernel: [ 3119.747191] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev rom.u inst 0 is active

Jan 6 10:48:49 ip kernel: [ 3119.747201] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev rom.u

Jan 6 10:48:49 ip kernel: [ 3119.747202] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev dma.xdma.u

Jan 6 10:48:49 ip kernel: [ 3119.747247] xocl:xdma_device_open: xocl device 0000:00:1d.0, 0x000000005fc88d0d.

Jan 6 10:48:49 ip kernel: [ 3119.747752] xocl:map_single_bar: map bar 0, len 33554432

Jan 6 10:48:49 ip kernel: [ 3119.747754] xocl:map_single_bar: map bar 1, len 2097152

Jan 6 10:48:49 ip kernel: [ 3119.747807] xocl:map_single_bar: BAR1 at 0x85400000 mapped at 0x00000000e79f298b, length=2097152(/2097152)

Jan 6 10:48:49 ip kernel: [ 3119.747810] xocl:map_single_bar: map bar 2, len 65536

Jan 6 10:48:49 ip kernel: [ 3119.747823] xocl:map_single_bar: BAR2 at 0x85600000 mapped at 0x00000000245e9658, length=65536(/65536)

Jan 6 10:48:49 ip kernel: [ 3119.747826] xocl:map_bars: config bar 2, pos 1.

Jan 6 10:48:49 ip kernel: [ 3119.747826] xocl:map_single_bar: map bar 3, len 0

Jan 6 10:48:49 ip kernel: [ 3119.747827] xocl:map_single_bar: map bar 4, len 137438953472

Jan 6 10:48:49 ip kernel: [ 3119.747828] xocl:map_single_bar: map bar 5, len 0

Jan 6 10:48:49 ip kernel: [ 3119.747829] xocl:identify_bars: xdev 0x000000008517ec9d, bars 2, config at 1.

Jan 6 10:48:49 ip kernel: [ 3119.747829] xocl:identify_bars: 2 BARs: config 1, user 1, bypass -1.

Jan 6 10:48:49 ip kernel: [ 3119.755625] xocl:enable_msi_msix: request vectors: h2c 4, c2h 4, user 16

Jan 6 10:48:49 ip kernel: [ 3119.755695] xocl:pci_keep_intx_enabled: 0000:00:1d.0: clear INTX_DISABLE, 0x406 -> 0x6.

Jan 6 10:48:49 ip kernel: [ 3119.755785] xocl:irq_msix_channel_setup: engine 0-H2C0-MM, irq#122.

Jan 6 10:48:49 ip kernel: [ 3119.755804] xocl:irq_msix_channel_setup: engine 0-H2C1-MM, irq#123.

Jan 6 10:48:49 ip kernel: [ 3119.755820] xocl:irq_msix_channel_setup: engine 0-H2C2-MM, irq#124.

Jan 6 10:48:49 ip kernel: [ 3119.755842] xocl:irq_msix_channel_setup: engine 0-H2C3-MM, irq#125.

Jan 6 10:48:49 ip kernel: [ 3119.755858] xocl:irq_msix_channel_setup: engine 0-C2H0-MM, irq#126.

Jan 6 10:48:49 ip kernel: [ 3119.755879] xocl:irq_msix_channel_setup: engine 0-C2H1-MM, irq#127.

Jan 6 10:48:49 ip kernel: [ 3119.755896] xocl:irq_msix_channel_setup: engine 0-C2H2-MM, irq#128.

Jan 6 10:48:49 ip kernel: [ 3119.755916] xocl:irq_msix_channel_setup: engine 0-C2H3-MM, irq#129.

Jan 6 10:48:49 ip kernel: [ 3119.755932] xocl:irq_msix_user_setup: 0-USR-0, IRQ#130 with 0x0000000083739a6d

Jan 6 10:48:49 ip kernel: [ 3119.755949] xocl:irq_msix_user_setup: 0-USR-1, IRQ#131 with 0x000000008ae17967

Jan 6 10:48:49 ip kernel: [ 3119.755970] xocl:irq_msix_user_setup: 0-USR-2, IRQ#132 with 0x000000005e54f2c5

Jan 6 10:48:49 ip kernel: [ 3119.755986] xocl:irq_msix_user_setup: 0-USR-3, IRQ#133 with 0x000000000ee8c643

Jan 6 10:48:49 ip kernel: [ 3119.756006] xocl:irq_msix_user_setup: 0-USR-4, IRQ#134 with 0x0000000039ae046f

Jan 6 10:48:49 ip kernel: [ 3119.756021] xocl:irq_msix_user_setup: 0-USR-5, IRQ#135 with 0x00000000b27b8661

Jan 6 10:48:49 ip kernel: [ 3119.756045] xocl:irq_msix_user_setup: 0-USR-6, IRQ#136 with 0x00000000d213e848

Jan 6 10:48:49 ip kernel: [ 3119.756060] xocl:irq_msix_user_setup: 0-USR-7, IRQ#137 with 0x00000000287f528b

Jan 6 10:48:49 ip kernel: [ 3119.756075] xocl:irq_msix_user_setup: 0-USR-8, IRQ#138 with 0x000000002ca52cbc

Jan 6 10:48:49 ip kernel: [ 3119.756095] xocl:irq_msix_user_setup: 0-USR-9, IRQ#139 with 0x0000000008836bf9

Jan 6 10:48:49 ip kernel: [ 3119.756110] xocl:irq_msix_user_setup: 0-USR-10, IRQ#140 with 0x000000001a1c86d2

Jan 6 10:48:49 ip kernel: [ 3119.756131] xocl:irq_msix_user_setup: 0-USR-11, IRQ#141 with 0x00000000daff9f21

Jan 6 10:48:49 ip kernel: [ 3119.756147] xocl:irq_msix_user_setup: 0-USR-12, IRQ#142 with 0x000000005b6f162f

Jan 6 10:48:49 ip kernel: [ 3119.756167] xocl:irq_msix_user_setup: 0-USR-13, IRQ#143 with 0x00000000a99095e8

Jan 6 10:48:49 ip kernel: [ 3119.756182] xocl:irq_msix_user_setup: 0-USR-14, IRQ#144 with 0x00000000a2557c42

Jan 6 10:48:49 ip kernel: [ 3119.756197] xocl:irq_msix_user_setup: 0-USR-15, IRQ#145 with 0x0000000050414470

Jan 6 10:48:49 ip kernel: [ 3119.756237] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev dma.xdma inst 2097152

Jan 6 10:48:49 ip kernel: [ 3119.756240] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev dma.xdma.u inst 2097152 is active

Jan 6 10:48:49 ip kernel: [ 3119.756242] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev mb_scheduler.u

Jan 6 10:48:49 ip kernel: [ 3119.756248] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x82180000-0x82180fff]

Jan 6 10:48:49 ip kernel: [ 3119.756249] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [mem 0x82190000-0x8219ffff]

Jan 6 10:48:49 ip kernel: [ 3119.756252] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: resource [irq 0-3]

Jan 6 10:48:49 ip kernel: [ 3119.756335] xocl_mb_sche mb_scheduler.u.5242880: dev ffff8c133c705010, exec_create: CQ size is 65536

Jan 6 10:48:49 ip kernel: [ 3119.756347] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec_reset(0) cfg(0)

Jan 6 10:48:49 ip kernel: [ 3119.756349] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec_reset resets

Jan 6 10:48:49 ip kernel: [ 3119.756352] xocl 0000:00:1d.0: dev ffff8c133e0340a0, exec_reset: exec->xclbin(00000000-0000-0000-0000-000000000000),xclbin(00000000-0000-0000-0000-000000000000)

Jan 6 10:48:49 ip kernel: [ 3119.756433] [drm] command scheduler started

Jan 6 10:48:49 ip kernel: [ 3119.756473] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev mb_scheduler inst 5242880

Jan 6 10:48:49 ip kernel: [ 3119.756476] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev mb_scheduler.u inst 5242880 is active

Jan 6 10:48:49 ip kernel: [ 3119.756478] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev mailbox.u

Jan 6 10:48:49 ip kernel: [ 3119.756795] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_probe: successfully initialized

Jan 6 10:48:49 ip kernel: [ 3119.756810] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev mailbox inst 15728640

Jan 6 10:48:49 ip kernel: [ 3119.756917] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev mailbox.u inst 15728640 is active

Jan 6 10:48:49 ip kernel: [ 3119.756919] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: creating subdev icap.u

Jan 6 10:48:49 ip kernel: [ 3119.756956] icap.u icap.u.16777216: dev ffff8c133e189410, icap_probe: successfully initialized FPGA IDCODE 0x0

Jan 6 10:48:49 ip kernel: [ 3119.756963] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: Created subdev icap inst 16777216

Jan 6 10:48:49 ip kernel: [ 3119.756965] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_create: subdev icap.u inst 16777216 is active

Jan 6 10:48:49 ip kernel: [ 3119.756969] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.

Jan 6 10:48:49 ip kernel: [ 3119.756970] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_subdev_vsec: No Vendor Specific Capability.

Jan 6 10:48:49 ip kernel: [ 3119.756972] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_p2p_init: Initializing P2P, bar 4, len 137438953472

Jan 6 10:48:49 ip kernel: [ 3119.757038] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 15 via SW

Jan 6 10:48:51 ip kernel: [ 3122.181787] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, timeout_msg: found outstanding msg time'd out

Jan 6 10:48:51 ip kernel: [ 3122.181789] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, timeout_msg: peer becomes dead

Jan 6 10:48:51 ip kernel: [ 3122.181823] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62

Jan 6 10:48:51 ip kernel: [ 3122.182191] [drm] Initialized xocl 2.5.0 20200103 for 0000:00:1d.0 on minor 0

Jan 6 10:48:51 ip kernel: [ 3122.182260] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 11 via SW

Jan 6 10:48:51 ip kernel: [ 3122.182263] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_connect: ch_state 0x0, ret -107

Jan 6 10:48:51 ip kernel: [ 3122.189798] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_refresh_subdevs: get fdt from peer

Jan 6 10:48:51 ip kernel: [ 3122.190203] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 10 via SW

Jan 6 10:49:22 ip systemd[1]: Started Xilinx Management Proxy Daemon (MPD).

Jan 6 10:49:22 ip mpd[1859]: started

Jan 6 10:49:22 ip mpd[1859]: found mpd plugin: /opt/xilinx/xrt/lib/libmpd_plugin.so

Jan 6 10:49:22 ip mpd[1859]: aws mpd plugin init called: 0

Jan 6 10:49:22 ip mpd[1859]: create thread pair for 0000:00:1d.0

Jan 6 10:49:22 ip mpd[1859]: 1 pairs of threads running...

Jan 6 10:49:22 ip mpd[1859]: [0:0:1d.0] write 56 bytes out of 56 bytes to fd 4

Jan 6 10:49:22 ip mpd[1859]: [0:0:1d.0] msg arrived on mailbox fd 4

Jan 6 10:49:22 ip mpd[1859]: [0:0:1d.0] retrieved msg size from mailbox: 40 bytes

Jan 6 10:49:22 ip mpd[1859]: [0:0:1d.0] read 72 bytes out of 72 bytes from fd 4, valid: 1

Jan 6 10:49:22 ip mpd[1859]: [0:0:1d.0] mpd daemon: request 11 received(reqSize: 24)

Jan 6 10:49:22 ip kernel: [ 3152.625936] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, _xocl_drvinst_open: OPEN 1

Jan 6 10:49:22 ip kernel: [ 3152.625999] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, dequeue_rx_msg: peer becomes active

Jan 6 10:49:22 ip kernel: [ 3152.626035] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, process_request: received request from peer: 12, passed on

Jan 6 10:49:22 ip kernel: [ 3152.626038] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mailbox_srv: received request (12) from peer

Jan 6 10:49:22 ip kernel: [ 3152.626040] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mailbox_srv: mgmt driver online

Jan 6 10:49:22 ip kernel: [ 3152.626103] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 11 via SW

Jan 6 10:49:22 ip kernel: [ 3152.626235] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_read: Software TX msg is too big

Jan 6 10:49:25 ip mpd[1859]: [0:0:1d.0] msg arrived on mailbox fd 4

Jan 6 10:49:25 ip mpd[1859]: [0:0:1d.0] retrieved msg size from mailbox: 32 bytes

Jan 6 10:49:25 ip mpd[1859]: [0:0:1d.0] read 64 bytes out of 64 bytes from fd 4, valid: 1

Jan 6 10:49:25 ip kernel: [ 3155.623617] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_request: sending request: 15 via SW

Jan 6 10:49:25 ip kernel: [ 3155.623742] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_read: Software TX msg is too big

Jan 6 10:49:46 ip kernel: [ 3176.649814] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, timeout_msg: found waiting msg time'd out

Jan 6 10:49:46 ip kernel: [ 3176.659006] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_connect: ch_state 0x0, ret -62

Jan 6 10:49:49 ip kernel: [ 3179.625808] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, timeout_msg: found waiting msg time'd out

Jan 6 10:49:49 ip kernel: [ 3179.635898] xocl 0000:00:1d.0: dev ffff8c133e0340a0, xocl_mb_read_p2p_addr: dropped request (15), failed with err: -62

Jan 6 10:49:49 ip kernel: [ 3179.635912] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev icap, cdev (null)

Jan 6 10:49:49 ip kernel: [ 3179.635935] icap.u icap.u.16777216: dev ffff8c133e189410, icap_remove: cleaned up successfully

Jan 6 10:49:49 ip kernel: [ 3179.635990] xocl 0000:00:1d.0: dev ffff8c133e0340a0, __xocl_subdev_destroy: Destroy subdev mailbox, cdev 0000000086a62c1b

Jan 6 10:49:49 ip kernel: [ 3179.636308] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, mailbox_remove: mailbox cleaned up successfully

Jan 6 10:49:49 ip kernel: [ 3179.636313] mailbox.u mailbox.u.15728640: dev ffff8c133c701010, xocl_drvinst_free: Wait for close 00000000067277db

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Xilinx/FPGA_as_a_Service/issues/7?email_source=notifications&email_token=ADFZBM2T5T5DL3ZE77B3UWTQ4MFB5A5CNFSM4J26WL3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEIFDZBQ#issuecomment-571096198, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADFZBMZSIIDRWKRD2K2PW3TQ4MFB5ANCNFSM4J26WL3A .

Thanks for off-line help, issue is solved!!

Xilinx / FPGA_as_a_Service

Try on AWS F1, but result looks not promised? #7

docker pull xilinxatg/aws_k8s_fpga_plugin:06272019

kubectl get ds -n kube-system

kubectl delete ds xxx -n kube-system

mount |grep sysfs

ls /sys/bus/pci/devices/0000\:00\:1d.0/drm/

您好，感谢建议。折腾了半天，我应该是把XRT安装起来了。（问题1：XRT没有安装，为什么aws AFI可以load并运行呢? 是否AWS F1机器上面是自己customized Runtime?)

systemctl status mpd

xbutil scan

systemctl status mpd

systemctl start mpd

您好，多谢。对了xclmgmt是Xilinx的工具。aws 是fpga-load/clear* 那一套工具是吗？我这台instance上面安装了aws cli，这个会有冲突吗？

systemctl stop mpd

rmmod xocl

modprobe xocl

systemctl start mpd

systemctl stop mpd

rmmod xocl

modprobe xocl

systemctl start mpd

Xilinx / FPGA_as_a_Service

Try on AWS F1, but result looks not promised? #7

docker pull xilinxatg/aws_k8s_fpga_plugin:06272019

kubectl get ds -n kube-system

kubectl delete ds xxx -n kube-system

mount |grep sysfs

ls /sys/bus/pci/devices/0000\:00\:1d.0/drm/

您好，感谢建议。 折腾了半天，我应该是把XRT安装起来了。（问题1：XRT没有安装，为什么aws AFI可以load并运行呢? 是否AWS F1机器上面是自己customized Runtime?)

systemctl status mpd

xbutil scan

systemctl status mpd

systemctl start mpd

您好，多谢。对了xclmgmt是Xilinx的工具。aws 是fpga-load/clear* 那一套工具是吗？ 我这台instance上面安装了aws cli，这个会有冲突吗？

systemctl stop mpd

rmmod xocl

modprobe xocl

systemctl start mpd

systemctl stop mpd

rmmod xocl

modprobe xocl

systemctl start mpd

您好，感谢建议。折腾了半天，我应该是把XRT安装起来了。（问题1：XRT没有安装，为什么aws AFI可以load并运行呢? 是否AWS F1机器上面是自己customized Runtime?)

您好，多谢。对了xclmgmt是Xilinx的工具。aws 是fpga-load/clear* 那一套工具是吗？我这台instance上面安装了aws cli，这个会有冲突吗？