karylab_agents/vllm
1
0
Fork 0
forked from Karylab-cklius/vllm
Code Pull Requests 2 Actions 1 Packages Activity

[ROCm][CI] move lora%N test to mi300 and gate (#45970)

Browse Source 
Signed-off-by: Divakar Verma <divakar.verma@amd.com>
This commit is contained in:
Divakar Verma
2026-06-19 00:50:32 +08:00
committed by GitHub
parent 6c379b9e54
commit 21da47dabe
2 changed files with 25 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.buildkite/test-amd.yaml
+14 -16
View File
@@ -415,22 +415,6 @@ steps:
commands:
- pytest -v -s kernels/mamba
#----------------------------------------------------------- mi250 · lora ------------------------------------------------------------#
- label: LoRA %N # TBD
timeout_in_minutes: 180
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
agent_pool: mi250_1
parallelism: 4
optional: true
working_dir: "/vllm-workspace/tests"
source_file_dependencies:
- vllm/lora
- tests/lora
- vllm/platforms/rocm.py
commands:
- pytest -v -s lora --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --ignore=lora/test_chatglm3_tp.py --ignore=lora/test_llama_tp.py --ignore=lora/test_qwen3_with_multi_loras.py --ignore=lora/test_olmoe_tp.py --ignore=lora/test_deepseekv2_tp.py --ignore=lora/test_gptoss_tp.py --ignore=lora/test_qwen3moe_tp.py --ignore=lora/test_qwen35_densemodel_lora.py
#------------------------------------------------------ mi250 · models / basic -------------------------------------------------------#
- label: Basic Models Test (Other CPU) # TBD
@@ -1699,6 +1683,20 @@ steps:
#----------------------------------------------------------- mi300 · lora ------------------------------------------------------------#
- label: LoRA %N # TBD
timeout_in_minutes: 180
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
agent_pool: mi300_1
parallelism: 4
optional: true
working_dir: "/vllm-workspace/tests"
source_file_dependencies:
- vllm/lora
- tests/lora
- vllm/platforms/rocm.py
commands:
- pytest -v -s lora --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --ignore=lora/test_chatglm3_tp.py --ignore=lora/test_llama_tp.py --ignore=lora/test_qwen3_with_multi_loras.py --ignore=lora/test_olmoe_tp.py --ignore=lora/test_deepseekv2_tp.py --ignore=lora/test_gptoss_tp.py --ignore=lora/test_qwen3moe_tp.py --ignore=lora/test_qwen35_densemodel_lora.py
- label: LoRA TP (Distributed) # TBD
timeout_in_minutes: 180
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
.buildkite/test_areas/lora.yaml
+11
View File
@@ -12,6 +12,17 @@ steps:
commands:
- pytest -v -s lora --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --ignore=lora/test_chatglm3_tp.py --ignore=lora/test_llama_tp.py --ignore=lora/test_qwen3_with_multi_loras.py --ignore=lora/test_olmoe_tp.py --ignore=lora/test_deepseekv2_tp.py --ignore=lora/test_gptoss_tp.py --ignore=lora/test_qwen3moe_tp.py --ignore=lora/test_qwen35_densemodel_lora.py
parallelism: 4
mirror:
amd:
device: mi325_1
working_dir: "/vllm-workspace/tests"
timeout_in_minutes: 60
source_file_dependencies:
- vllm/lora
- tests/lora
- vllm/platforms/rocm.py
depends_on:
- image-build-amd
- label: LoRA TP (Distributed)