Skip to content

Fix: compile host_build_graph orchestration with host g++ #1330

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

Merged
ChaoWao merged 1 commit into from
Jul 11, 2026
Merged

Fix: compile host_build_graph orchestration with host g++ #1330

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
30 changes: 18 additions & 12 deletions simpler_setup/kernel_compiler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -419,25 +419,31 @@ def compile_orchestration(
if orch_includes:
include_dirs = include_dirs + orch_includes

# Resolve toolchain: HOST_GXX needs no runtime-specific extras
toolchain_type = self._get_toolchain(
{
"a2a3": ToolchainType.AARCH64_GXX,
"a2a3sim": ToolchainType.HOST_GXX,
"a5": ToolchainType.AARCH64_GXX,
"a5sim": ToolchainType.HOST_GXX,
},
)
# host_build_graph dlopens the orchestration .so on the host, so it
# compiles with the host g++ (x86_64) regardless of platform.
# tensormap_and_ringbuffer dlopens it on the aarch64 AICPU onboard, so
# it cross-compiles there and uses the host g++ only for sim.
if runtime_name == "host_build_graph":
toolchain_type = ToolchainType.HOST_GXX
elif runtime_name == "tensormap_and_ringbuffer":
toolchain_type = self._get_toolchain(
{
"a2a3": ToolchainType.AARCH64_GXX,
"a2a3sim": ToolchainType.HOST_GXX,
"a5": ToolchainType.AARCH64_GXX,
"a5sim": ToolchainType.HOST_GXX,
},
)
Comment thread
ChaoWao marked this conversation as resolved.
else:
raise ValueError(f"Unknown runtime_name: {runtime_name!r}")
toolchain: Union[GxxToolchain, Aarch64GxxToolchain]
if toolchain_type == ToolchainType.AARCH64_GXX:
assert self.aarch64 is not None, "aarch64 toolchain is only available for hardware platforms"
toolchain = self.aarch64
else:
toolchain = self.host_gxx

# HOST_GXX: simulation build (host execution)
# AARCH64_GXX: cross-compilation for supported runtimes
# Note: orchestration uses ops table via pto_orchestration_api.h (no extra runtime sources needed)
# Orchestration uses the ops table via pto_orchestration_api.h (no extra runtime sources needed).
return self._compile_orchestration_shared_lib(
source_path,
toolchain,
Expand Down
Loading