Make ORT_MIGRAPHX_MODEL_CACHE_PATH unconditional in packaged systemd unit

Without a valid MIGraphX cache dir, ONNX Runtime's MIGraphX EP reads an
uninitialized cache path and crashes on the first query rather than just
recompiling on every restart -- confirmed by hand while verifying the rocm
backend. Previously commented out as an opt-in perf tweak; it's actually
required for the rocm backend to work at all, so bakery-installed systems
need it set by default. No-op for cpu/npu/cuda backends.

Version bump: breadmill 0.2.1 -> 0.2.2.
This commit is contained in:
Breadway 2026-07-03 22:38:41 +08:00
parent fdf596e58a
commit 3b709b3a77
3 changed files with 7 additions and 6 deletions

2
Cargo.lock generated
View file

@ -133,7 +133,7 @@ dependencies = [
[[package]]
name = "breadmill"
version = "0.2.1"
version = "0.2.2"
dependencies = [
"breadsearch-shared",
"hex",

View file

@ -1,6 +1,6 @@
[package]
name = "breadmill"
version = "0.2.1"
version = "0.2.2"
edition = "2021"
license = "MIT"

View file

@ -5,10 +5,11 @@ After=default.target
[Service]
Type=simple
# Uncomment if built with --features rocm: persists MIGraphX's compiled-kernel
# cache across restarts (each new sequence length otherwise costs a ~60-120s
# recompile). See README.md#gpu-backend-notes.
#Environment=ORT_MIGRAPHX_MODEL_CACHE_PATH=%h/.cache/breadsearch/migraphx-cache
# Required (not just a perf tweak) when backend = "rocm": without a valid
# cache dir, ONNX Runtime's MIGraphX EP reads an uninitialized cache path and
# crashes on the first query instead of just recompiling every restart. Inert
# for cpu/npu/cuda backends. See README.md#gpu-backend-notes.
Environment=ORT_MIGRAPHX_MODEL_CACHE_PATH=%h/.cache/breadsearch/migraphx-cache
ExecStart=%h/.cargo/bin/breadmill
Restart=on-failure
RestartSec=5