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 9, 18:15 UTC
[CI][Attention] Add more CI dependencies for attention tests (#32487)
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Matthew Bonanni committed
Jan 22, 2026 at 13:44 UTC
300622e609a52b76526182bdfca615d5f9e448a2
3 files changed
+12
.buildkite/test-amd.yaml
+4
@@ -430,6 +430,8 @@ steps:
430
timeout_in_minutes: 30
431
gpu: h100
432
source_file_dependencies:
433
+ - vllm/config/attention.py
434
+ - vllm/model_executor/layers/attention
435
- vllm/v1/attention
436
- tests/v1/attention
437
commands:
@@ -454,6 +456,8 @@ steps:
456
timeout_in_minutes: 30
457
gpu: b200
458
source_file_dependencies:
459
+ - vllm/config/attention.py
460
+ - vllm/model_executor/layers/attention
461
- vllm/v1/attention
462
- tests/v1/attention
463
commands:
.buildkite/test-pipeline.yaml
+4
@@ -376,6 +376,8 @@ steps:
376
timeout_in_minutes: 30
377
gpu: h100
378
source_file_dependencies:
379
+ - vllm/config/attention.py
380
+ - vllm/model_executor/layers/attention
381
- vllm/v1/attention
382
- tests/v1/attention
383
commands:
@@ -398,6 +400,8 @@ steps:
400
timeout_in_minutes: 30
401
gpu: b200
402
source_file_dependencies:
403
+ - vllm/config/attention.py
404
+ - vllm/model_executor/layers/attention
405
- vllm/v1/attention
406
- tests/v1/attention
407
commands:
.buildkite/test_areas/attention.yaml
+4
@@ -6,6 +6,8 @@ steps:
6
timeout_in_minutes: 30
7
gpu: h100
8
source_file_dependencies:
9
+ - vllm/config/attention.py
10
+ - vllm/model_executor/layers/attention
11
- vllm/v1/attention
12
- tests/v1/attention
13
commands:
@@ -15,6 +17,8 @@ steps:
17
timeout_in_minutes: 30
18
gpu: b200
19
source_file_dependencies:
20
+ - vllm/config/attention.py
21
+ - vllm/model_executor/layers/attention
22
- vllm/v1/attention
23
- tests/v1/attention
24
commands: