Skip to content

Refactor and consolidate the TorchTPU KVCacheStore end-to-end test suite for cleaner abstractions. #798

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
copybara-service wants to merge 1 commit into from
Closed

Refactor and consolidate the TorchTPU KVCacheStore end-to-end test suite for cleaner abstractions. #798

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions run_tests.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,9 +64,11 @@ fi
if [ "$RUN_TORCH" = true ]; then
echo "=== Running Torch Python Unit Tests ==="
python "${WORKSPACE_DIR}/tpu_sync/frameworks/torch/kv_cache_manager_test.py"
python "${WORKSPACE_DIR}/tpu_sync/api/torch/kv_cache_manager_host_test.py"
python "${WORKSPACE_DIR}/tpu_sync/api/torch/kv_cache_manager_test.py"
python "${WORKSPACE_DIR}/tpu_sync/api/torch/kv_cache_store_test.py"
python "${WORKSPACE_DIR}/tpu_sync/api/torch/kv_cache_manager_transfer_test.py"
python "${WORKSPACE_DIR}/tpu_sync/api/torch/weight_synchronizer_test.py"
python "${WORKSPACE_DIR}/tpu_sync/api/torch/kv_cache_manager_mpmd_test.py"
python "${WORKSPACE_DIR}/tpu_sync/api/torch/kv_cache_store_recovery_e2e_test.py"
fi
10 changes: 4 additions & 6 deletions tpu_sync/api/torch/BUILD
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -165,16 +165,14 @@ py_test(
],
)

# Base accelerator test requiring physical TPU devices via torch_tpu.
# Tagged "manual" because generic CPU Forge runners lack TPU hardware.
# For remote execution on Forge, use the _gf (TPU v4) or _gl (TPU v5e) targets below.
py_test(
name = "kv_cache_store_recovery_e2e_test",
srcs = ["kv_cache_store_recovery_e2e_test.py"],
tags = [
"cpu:8",
"nobuilder",
"notap",
"requires-accel-hbm",
"requires-ghostlite:8",
"requires-net:external",
"manual",
],
deps = [
":kv_cache_manager_torch_py",
Expand Down
Loading
Loading