ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

install: keep pinned repairs off UV_TORCH_BACKEND and narrow setup.ps1's rocm pin match

Browse source 
Round 2 of the pinned-index hardening:

1. _build_uv_cmd converted UV_TORCH_BACKEND into --torch-backend before the
   new env isolation could act, and uv's torch backend redirects torch
   resolution to its own per-backend index even when --index-url is given
   (verified: a cu128-pinned dry run with UV_TORCH_BACKEND=cpu resolves
   torch 2.13.0+cpu). Pinned-index commands now never receive the flag and
   UV_TORCH_BACKEND joins the stripped env vars, so uv cannot re-read it.

2. setup.ps1's pinned reroute had the same bare rocm* glob install.ps1 had:
   a custom find-links leaf like rocm-rel-7.2.1 was routed through the ROCm
   --index-url path instead of the verbatim unknown-pin path. Now requires
   a digit after rocm, matching install.ps1, install.sh and
   _is_pip_rocm_family_leaf.

3. The marker test's case-normalization checks used -eq, which is
   case-insensitive in PowerShell, making them vacuous, and the unknown-leaf
   expectation was written lowercased while the implementation deliberately
   preserves custom-leaf case. Tightened to -ceq with the case-preserving
   expected value.

Adds unit + parity tests for 1 and 2 (5 new tests).
This commit is contained in:
Daniel Han 2026-07-12 10:38:38 +00:00
commit f8bde5e387
5 changed files with 76 additions and 13 deletions
Download patch file Download diff file Expand all files Collapse all files

34
studio/install_python_stack.py
View file

@ -2560,9 +2560,13 @@ def _build_uv_cmd(args: tuple[str, ...]) -> list[str]:
cmd.extend(_translate_pip_args_for_uv(args))
# Torch is pre-installed by install.sh/setup.ps1. Do not add
# --torch-backend by default -- it can cause solver dead-ends on CPU-only
# machines. Callers that need it can set UV_TORCH_BACKEND.
# machines. Callers that need it can set UV_TORCH_BACKEND. Never add it to a
# pinned-index command: uv's torch backend redirects torch resolution to its
# own per-backend index (verified: --index-url .../cu128 with UV_TORCH_BACKEND
# =cpu resolves torch+cpu), which would defeat the explicit pin exactly like
# the index env vars _install_env_for_cmd() strips.
_tb = os.environ.get("UV_TORCH_BACKEND", "")
if _tb:
if _tb and not _is_pinned_index_cmd(cmd):
cmd.append(f"--torch-backend={_tb}")
return cmd
@ -2576,7 +2580,22 @@ def _build_uv_cmd(args: tuple[str, ...]) -> list[str]:
# actually used. install.sh (run_install_cmd, #6898), install.ps1 and setup.ps1
# already neutralise these vars for their pinned installs; do the same here so the
# pin wins on every platform. Non-pinned installs (no --index-url) keep the mirror.
_UV_INDEX_ENV_VARS = ("UV_DEFAULT_INDEX", "UV_INDEX_URL", "UV_INDEX", "UV_EXTRA_INDEX_URL")
# UV_TORCH_BACKEND is stripped too: uv's torch backend redirects torch resolution
# to its own per-backend index even when a --index-url is given, so it defeats a
# pin the same way ( _build_uv_cmd also refuses to turn it into a --torch-backend
# flag for pinned commands; this covers uv reading the env var directly).
_UV_INDEX_ENV_VARS = (
"UV_DEFAULT_INDEX",
"UV_INDEX_URL",
"UV_INDEX",
"UV_EXTRA_INDEX_URL",
"UV_TORCH_BACKEND",
)
def _is_pinned_index_cmd(cmd: "list[str] | tuple[str, ...]") -> bool:
"""True when the command pins an index via --index-url / --default-index."""
return any(arg in ("--index-url", "--default-index") for arg in cmd)
def _install_env_for_cmd(cmd: "list[str]") -> "dict[str, str] | None":
@ -2585,11 +2604,12 @@ def _install_env_for_cmd(cmd: "list[str]") -> "dict[str, str] | None":
Returns None (inherit the caller's environment unchanged) when the command
does NOT pin an index, so ordinary installs still honour a user's UV_INDEX /
UV_EXTRA_INDEX_URL mirror. When the command passes --index-url / --default-index
(the torch repair paths), the four uv index env vars are removed so the pinned
index is not overridden by an inherited mirror (uv treats the default index as
lowest priority). Mirrors install.sh's run_install_cmd gate (#6898).
(the torch repair paths), the uv index env vars (and UV_TORCH_BACKEND) are
removed so the pinned index is not overridden by an inherited mirror or torch
backend (uv treats the default index as lowest priority). Mirrors install.sh's
run_install_cmd gate (#6898).
"""
if not any(arg in ("--index-url", "--default-index") for arg in cmd):
if not _is_pinned_index_cmd(cmd):
return None
env = os.environ.copy()
for name in _UV_INDEX_ENV_VARS: