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 16, 23:15 UTC
[ci] Add AMD, Neuron, Intel tests for AWS CI and turn off default soft fail for GPU tests (#5464)
Signed-off-by: kevin <kevin@anyscale.com>
Kevin H. Luu committed
Jun 12, 2024 at 14:00 UTC
8b82a89997826af8e0e4ecfaaed60f3b28b1baed
1 file changed
+29
-1
.buildkite/test-template-aws.j2
+29
-1
@@ -19,6 +19,34 @@ steps:
19
limit: 5
20
- wait
21
22
+ - group: "AMD Tests"
23
+ depends_on: ~
24
+ steps:
25
+ {% for step in steps %}
26
+ {% if step.mirror_hardwares and "amd" in step.mirror_hardwares %}
27
+ - label: "AMD: {{ step.label }}"
28
+ agents:
29
+ queue: amd
30
+ command: bash .buildkite/run-amd-test.sh "cd {{ (step.working_dir or default_working_dir) | safe }} ; {{ step.command or (step.commands | join(" ; ")) | safe }}"
31
+ env:
32
+ DOCKER_BUILDKIT: "1"
33
+ soft_fail: true
34
+ {% endif %}
35
+ {% endfor %}
36
+
37
+ - label: "Neuron Test"
38
+ depends_on: ~
39
+ agents:
40
+ queue: neuron
41
+ command: bash .buildkite/run-neuron-test.sh
42
+ soft_fail: false
43
+
44
+ - label: "Intel Test"
45
+ depends_on: ~
46
+ agents:
47
+ queue: intel
48
+ command: bash .buildkite/run-cpu-test.sh
49
+
50
{% for step in steps %}
51
- label: "{{ step.label }}"
52
agents:
@@ -31,7 +59,7 @@ steps:
59
{% else %}
60
queue: gpu_1_queue
61
{% endif %}
34
- soft_fail: true
62
+ soft_fail: {{ step.soft_fail or false }}
63
{% if step.parallelism %}
64
parallelism: {{ step.parallelism }}
65
{% endif %}