From f2cc67750f55a71edff0c527a1bfc14fde8132c3 Mon Sep 17 00:00:00 2001
From: Mamatha Inamdar <mamatha4@linux.vnet.ibm.com>
Date: Mon, 8 Nov 2021 10:50:03 +0530
Subject: [PATCH] [nvidia]:Patch to update nvidia plugin for GPU info
This patch is to update nvidia plugin to collect
logs for Nvidia GPUs
Signed-off-by: Mamatha Inamdar <mamatha4@linux.vnet.ibm.com>
Reported-by: Borislav Stoymirski <borislav.stoymirski@bg.ibm.com>
Reported-by: Yesenia Jimenez <yesenia@us.ibm.com>
---
sos/report/plugins/nvidia.py | 15 +++++++++++++--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/sos/report/plugins/nvidia.py b/sos/report/plugins/nvidia.py
index 09aaf586b..9e21b478e 100644
--- a/sos/report/plugins/nvidia.py
+++ b/sos/report/plugins/nvidia.py
@@ -23,13 +23,24 @@ def setup(self):
'--list-gpus',
'-q -d PERFORMANCE',
'-q -d SUPPORTED_CLOCKS',
- '-q -d PAGE_RETIREMENT'
+ '-q -d PAGE_RETIREMENT',
+ '-q',
+ '-q -d ECC',
+ 'nvlink -s',
+ 'nvlink -e'
]
self.add_cmd_output(["nvidia-smi %s" % cmd for cmd in subcmds])
query = ('gpu_name,gpu_bus_id,vbios_version,temperature.gpu,'
- 'utilization.gpu,memory.total,memory.free,memory.used')
+ 'utilization.gpu,memory.total,memory.free,memory.used,'
+ 'clocks.applications.graphics,clocks.applications.memory')
+ querypages = ('timestamp,gpu_bus_id,gpu_serial,gpu_uuid,'
+ 'retired_pages.address,retired_pages.cause')
self.add_cmd_output("nvidia-smi --query-gpu=%s --format=csv" % query)
+ self.add_cmd_output(
+ "nvidia-smi --query-retired-pages=%s --format=csv" % querypages
+ )
+ self.add_journal(boot=0, identifier='nvidia-persistenced')
# vim: set et ts=4 sw=4 :