A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 17, 03:00 UTC
[CI] Move CPU/AMD tests to after wait (#4123)
Cade Daniel committed
Apr 16, 2024 at 22:53 UTC
11d652bd4f6f81d09638399885099b78a4e3b9c8
1 file changed
+8
-7
.buildkite/test-template.j2
+8
-7
@@ -3,13 +3,6 @@
3
{% set default_working_dir = "/vllm-workspace/tests" %}
4
5
steps:
6
- - label: "AMD Test"
7
- agents:
8
- queue: amd
9
- command: bash .buildkite/run-amd-test.sh
10
-
11
- - label: "CPU Test"
12
- command: bash .buildkite/run-cpu-test.sh
6
7
- label: ":docker: build image"
8
commands:
@@ -23,6 +16,14 @@ steps:
16
limit: 5
17
- wait
18
19
+ - label: "AMD Test"
20
+ agents:
21
+ queue: amd
22
+ command: bash .buildkite/run-amd-test.sh
23
+
24
+ - label: "CPU Test"
25
+ command: bash .buildkite/run-cpu-test.sh
26
+
27
{% for step in steps %}
28
- label: "{{ step.label }}"
29
agents: