Skip to content

[XLA:GPU] Derive replica and partition counts from HloModule in ExecuteReplicated. #34568

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

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

[XLA:GPU] Derive replica and partition counts from HloModule in ExecuteReplicated. #34568

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
6 changes: 4 additions & 2 deletions xla/tests/collective_ops_e2e_test_base.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -180,8 +180,10 @@ CollectiveOpsE2ETestBase::ExecuteReplicated(
absl::StatusOr<CollectiveOpsE2ETestBase::ExecutionResult>
CollectiveOpsE2ETestBase::ExecuteReplicated(
std::unique_ptr<HloModule> module,
const std::vector<std::vector<Literal*>> arguments, int64_t num_replicas,
int64_t num_partitions, bool run_hlo_passes) {
const std::vector<std::vector<Literal*>> arguments, bool run_hlo_passes) {
int64_t num_replicas = module->config().replica_count();
int64_t num_partitions = module->config().num_partitions();

CHECK(num_replicas > 0 && "expect at least one replica");
CHECK(num_partitions > 0 && "expect at least one partition");
CHECK(num_replicas == arguments.size() &&
Expand Down
3 changes: 1 addition & 2 deletions xla/tests/collective_ops_e2e_test_base.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -77,8 +77,7 @@ class CollectiveOpsE2ETestBase : public HloHardwareIndependentTestBase {

absl::StatusOr<ExecutionResult> ExecuteReplicated(
std::unique_ptr<HloModule> module,
std::vector<std::vector<Literal*>> arguments, int64_t num_replicas,
int64_t num_partitions, bool run_hlo_passes = true);
std::vector<std::vector<Literal*>> arguments, bool run_hlo_passes = true);

const se::GpuComputeCapability& Capability() {
return hlo_runner_->backend()
Expand Down
Loading
Loading