From bba50008f107d838744569b58b255039d30e7a7e Mon Sep 17 00:00:00 2001
From: Simo Tuomisto <simo.tuomisto@aalto.fi>
Date: Mon, 7 Sep 2020 15:20:48 +0300
Subject: [PATCH] Add functionality for enabling nvidia performance counters

---
 defaults/main.yml        | 2 ++
 tasks/main.yml           | 7 +++++++
 templates/nvidia.conf.j2 | 1 +
 3 files changed, 10 insertions(+)
 create mode 100644 templates/nvidia.conf.j2

diff --git a/defaults/main.yml b/defaults/main.yml
index a09b830..95d35a3 100644
--- a/defaults/main.yml
+++ b/defaults/main.yml
@@ -20,4 +20,6 @@ cuda_bash_cuda_root: "/usr/local/cuda"
 cuda_bash_cuda_inc_dir: "/usr/local/cuda/bin"
 cuda_bash_cpath: "/usr/local/cuda/include"
 
+cuda_enable_perf_counters: False
+
 # vim:ft=ansible:
diff --git a/tasks/main.yml b/tasks/main.yml
index 64e4adb..28625de 100644
--- a/tasks/main.yml
+++ b/tasks/main.yml
@@ -35,6 +35,13 @@
   - include_tasks: cuda_init.yml
     when: cuda_init
 
+  - name: Enable performance counters for all users via modprobe
+    template:
+      src: nvidia.conf.j2
+      dest: /etc/modprobe.d/nvidia.conf
+      mode: 0644
+    when: cuda_enable_perf_counters
+
   # This is here because if we in the same playbook try to start slurmd without
   # having run the cuda_init.sh script then slurmd doesn't start and the play fails.
   # todo: reload nvidia modules/etc instead of restart
diff --git a/templates/nvidia.conf.j2 b/templates/nvidia.conf.j2
new file mode 100644
index 0000000..aa6e3f4
--- /dev/null
+++ b/templates/nvidia.conf.j2
@@ -0,0 +1 @@
+options nvidia "NVreg_RestrictProfilingToAdminUsers=0"