From 97fe2bbf496f26fbf7a5052750dd782cbe779a48 Mon Sep 17 00:00:00 2001 From: vbedida79 Date: Tue, 17 Dec 2024 08:51:39 -0800 Subject: [PATCH] tests_gaudi: Add Intel Gaudi firmware check workload Signed-off-by: vbedida79 --- tests/gaudi/l2/README.md | 20 +++++++++++++++++++- tests/gaudi/l2/hl-smi-firmware_job.yaml | 18 ++++++++++++++++++ 2 files changed, 37 insertions(+), 1 deletion(-) create mode 100644 tests/gaudi/l2/hl-smi-firmware_job.yaml diff --git a/tests/gaudi/l2/README.md b/tests/gaudi/l2/README.md index 7f91c376..99195c1a 100644 --- a/tests/gaudi/l2/README.md +++ b/tests/gaudi/l2/README.md @@ -172,4 +172,22 @@ sh-5.1# curl http://vllm-workload.gaudi-validation.svc.cluster.local:8000/v1/com "max_tokens": 10 }' {"id":"cmpl-9a0442d0da67411081837a3a32a354f2","object":"text_completion","created":1730321284,"model":"meta-llama/Llama-3.1-8B","choices":[{"index":0,"text":" group of individual stars that forms a pattern or figure","logprobs":null,"finish_reason":"length","stop_reason":null}],"usage":{"prompt_tokens":5,"total_tokens":15,"completion_tokens":10}} -``` \ No newline at end of file +``` + +## Check firmware version with hl-smi +System Management Interface Tool (hl-smi) utility tool obtains information and monitors data of the Intel Gaudi AI accelerators. +Run below command to check firmware version with the tool: +``` +$ oc apply -f https://raw.githubusercontent.com/intel/intel-technology-enabling-for-openshift/main/tests/gaudi/l2/hl-smi-firmware_job.yaml +``` + +Verify Output: +``` +$ oc get pods -n gaudi-validation +NAME READY STATUS RESTARTS AGE +hl-smi-firmware-pxhsn 0/1 Completed 0 11s +``` +``` +$ oc logs hl-smi-firmware-pxhsn -n gaudi-validation + Firmware [SPI] Version : Preboot version hl-gaudi2-1.16.0-fw-50.1.2-sec-9 (May 26 2024 - 11:33:04) +``` diff --git a/tests/gaudi/l2/hl-smi-firmware_job.yaml b/tests/gaudi/l2/hl-smi-firmware_job.yaml new file mode 100644 index 00000000..f90c5230 --- /dev/null +++ b/tests/gaudi/l2/hl-smi-firmware_job.yaml @@ -0,0 +1,18 @@ +apiVersion: batch/v1 +kind: Job +metadata: + name: hl-smi-firmware + namespace: gaudi-validation +spec: + template: + metadata: + spec: + restartPolicy: Never + containers: + - name: hl-smi-firmware + image: vault.habana.ai/gaudi-docker/1.18.0/rhel9.4/habanalabs/pytorch-installer-2.4.0:1.18.0-524 + command: ["/bin/bash", "-c", "hl-smi -L | grep SPI"] + resources: + limits: + habana.ai/gaudi: 1 + imagePullPolicy: IfNotPresent