4123190b82
Both surfaced only once the staging matrix ran these suites on runners the
org queue does not cover. Neither is a product defect; both are tests
asserting something their environment cannot supply.
test_unsloth_pip_shim.py::test_forwarded_install_carries_protected_constraints
reads the ambient environment through importlib.metadata.distributions.
_protected_constraints_file correctly returns None when no protected
package is installed, so no --constraint pair is appended, and the test
then indexed execd[-2] unconditionally:
E IndexError: list index out of range
1 failed, 115 passed, 2 skipped
It failed on all four docker-test legs and in any bare venv, and passed
upstream only because studio-backend-ci installs torch and transformers
first. Its own sibling at line 93 already guards with len(execd) >= 2.
Guarding the index alone would have left the test measuring whatever
happened to be installed, so distributions() is now stubbed and the test
asserts the real contract deterministically. A second case covers the
other half of that contract, which is what a bare venv actually hits: with
nothing protected installed the install must still be forwarded, just
without the pair.
test_select_cuda_jit_tools.sh stages libnvrtc as symlinks and asserts
through readlink, because retargeting that symlink is what the function
under test does. git-bash copies instead of symlinking unless
MSYS=winsymlinks:nativestrict and the user is elevated, so readlink comes
back empty and all 14 assertions fail on both Windows runners, taking
tests/run_all.sh down with them for any Windows contributor. The code only
ever runs inside a Linux container, so probe for real symlink support and
skip when it is absent rather than assert something the filesystem cannot
represent.
Verified: the shim suite is 87 passed / 2 skipped in both a bare venv and
a full one; the shell suite still reports 14 passed on Linux and skips
with exit 0 under a simulated no-symlink filesystem.
122 lines
6.3 KiB
Bash
Executable file
122 lines
6.3 KiB
Bash
Executable file
#!/bin/bash
|
|
# SPDX-License-Identifier: AGPL-3.0-only
|
|
# Copyright 2026-present the Unsloth AI Inc. team. All rights reserved. See /studio/LICENSE.AGPL-3.0
|
|
# Unit tests for select_cuda_jit_tools() from docker/entrypoint.sh.
|
|
#
|
|
# cu12.8 is the immutable baked default; the cu13 tools switch on ONLY for sm_103
|
|
# and sm_121 (>= 580 drivers). The function picks per device via nvidia-smi
|
|
# compute_cap: those two arches retarget libnvrtc.so.12 -> the .cu13 alias (and
|
|
# point Triton at cu13 ptxas); every other arch keeps cu12.8 and leaves ptxas unset.
|
|
set -e
|
|
|
|
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
|
|
ENTRYPOINT_SH="$SCRIPT_DIR/../../docker/entrypoint.sh"
|
|
PASS=0
|
|
FAIL=0
|
|
|
|
# The fixtures below stage libnvrtc as symlinks and assert through readlink,
|
|
# because retargeting that symlink is exactly what the function under test does.
|
|
# git-bash copies instead of symlinking unless MSYS=winsymlinks:nativestrict and
|
|
# the user is elevated, so readlink comes back empty and all 14 assertions fail
|
|
# for reasons that have nothing to do with the code. That code only ever runs
|
|
# inside a Linux container, so skip rather than pretend: an unconditional run
|
|
# breaks tests/run_all.sh for every Windows contributor.
|
|
_probe=$(mktemp -d)
|
|
: > "$_probe/target"
|
|
if ! ln -s target "$_probe/link" 2>/dev/null || [ "$(readlink "$_probe/link")" != "target" ]; then
|
|
rm -rf "$_probe"
|
|
echo "=== test_select_cuda_jit_tools ==="
|
|
echo " SKIP: this filesystem does not honour symlinks (readlink cannot observe them)"
|
|
echo "PASS=0 FAIL=0 SKIPPED"
|
|
exit 0
|
|
fi
|
|
rm -rf "$_probe"
|
|
|
|
# Extract just the helper function (same sed range as the other function tests).
|
|
_FUNC_FILE=$(mktemp)
|
|
sed -n '/^select_cuda_jit_tools()/,/^}/p' "$ENTRYPOINT_SH" > "$_FUNC_FILE"
|
|
|
|
assert_eq() {
|
|
_label="$1"; _expected="$2"; _actual="$3"
|
|
if [ "$_actual" = "$_expected" ]; then
|
|
echo " PASS: $_label"
|
|
PASS=$((PASS + 1))
|
|
else
|
|
echo " FAIL: $_label (expected '$_expected', got '$_actual')"
|
|
FAIL=$((FAIL + 1))
|
|
fi
|
|
}
|
|
|
|
# $1 = compute_cap(s) the mock nvidia-smi reports, ONE PER LINE ("none" -> no
|
|
# nvidia-smi; multi-line models a mixed-GPU host). $2 (optional) = the initial
|
|
# libnvrtc.so.12 target (default cu12.8; "libnvrtc.so.12.cu13" models a stale
|
|
# link). Builds a fake Studio venv NVRTC dir. Prints "<PTXAS_STATE> <NVRTC_TARGET>".
|
|
run_select() {
|
|
_cap="$1"
|
|
_init="${2:-libnvrtc.so.12.cu128.orig}"
|
|
_tmp=$(mktemp -d)
|
|
mkdir -p "$_tmp/bin"
|
|
if [ "$_cap" != "none" ]; then
|
|
# nvidia-smi --query-gpu=compute_cap prints one cap per line; cat a file
|
|
# so an embedded newline in $_cap survives into the mock's output.
|
|
printf '%s\n' "$_cap" > "$_tmp/caps.txt"
|
|
printf '#!/bin/sh\ncat "%s"\n' "$_tmp/caps.txt" > "$_tmp/bin/nvidia-smi"
|
|
chmod +x "$_tmp/bin/nvidia-smi"
|
|
fi
|
|
_nvrtc="$_tmp/studio/unsloth_studio/lib/python3.12/site-packages/nvidia/cuda_nvrtc/lib"
|
|
mkdir -p "$_nvrtc"
|
|
: > "$_nvrtc/libnvrtc.so.12.cu128.orig" # real cu12.8 lib
|
|
: > "$_nvrtc/libnvrtc.so.13.stub" # stand-in cu13 lib
|
|
ln -sf libnvrtc.so.13.stub "$_nvrtc/libnvrtc.so.12.cu13" # staged cu13 alias
|
|
ln -sf "$_init" "$_nvrtc/libnvrtc.so.12" # cu12.8 default (or stale cu13)
|
|
bash -c '
|
|
set -euo pipefail
|
|
export PATH="'"$_tmp"'/bin:/usr/bin:/bin"
|
|
export UNSLOTH_STUDIO_HOME="'"$_tmp"'/studio"
|
|
unset TRITON_PTXAS_PATH || true
|
|
. "'"$_FUNC_FILE"'"
|
|
select_cuda_jit_tools || true
|
|
printf "%s %s\n" "${TRITON_PTXAS_PATH:-UNSET}" "$(readlink "'"$_nvrtc"'/libnvrtc.so.12")"
|
|
'
|
|
rm -rf "$_tmp"
|
|
}
|
|
|
|
echo "=== test_select_cuda_jit_tools ==="
|
|
|
|
# Non-DC arches: cu12.8 default is left untouched (no write) and ptxas unset
|
|
# (Triton keeps its bundled cu12.8 ptxas), so a 570-579 driver host -- root or
|
|
# --user -- is unaffected.
|
|
assert_eq "sm_80 Ampere -> cu128 default kept" "UNSET libnvrtc.so.12.cu128.orig" "$(run_select 8.0)"
|
|
assert_eq "sm_90 Hopper -> cu128 default kept" "UNSET libnvrtc.so.12.cu128.orig" "$(run_select 9.0)"
|
|
assert_eq "sm_100 B200 -> cu128 default kept" "UNSET libnvrtc.so.12.cu128.orig" "$(run_select 10.0)"
|
|
assert_eq "sm_120 RTX50 -> cu128 default kept" "UNSET libnvrtc.so.12.cu128.orig" "$(run_select 12.0)"
|
|
assert_eq "no nvidia-smi -> cu128 default kept" "UNSET libnvrtc.so.12.cu128.orig" "$(run_select none)"
|
|
|
|
# Blackwell datacenter: retarget libnvrtc.so.12 -> the .cu13 alias. ptxas stays
|
|
# UNSET here only because the test host has no /usr/local/cuda-13.0/bin/ptxas;
|
|
# the assertion that matters is that the NVRTC switched to cu13 for these arches.
|
|
assert_eq "sm_103 B300 -> cu13 NVRTC selected" "UNSET libnvrtc.so.12.cu13" "$(run_select 10.3)"
|
|
assert_eq "sm_121 DGX Spark -> cu13 NVRTC selected" "UNSET libnvrtc.so.12.cu13" "$(run_select 12.1)"
|
|
|
|
# Mixed-GPU hosts: a datacenter Blackwell (sm_103 / sm_121) sitting BEHIND an
|
|
# H100/B200 in the nvidia-smi ordering must still switch to cu13 -- every visible
|
|
# cap is scanned, not just the first. A host with no datacenter Blackwell at all
|
|
# keeps the cu12.8 default regardless of order.
|
|
assert_eq "H100 then B300 -> cu13 NVRTC selected" "UNSET libnvrtc.so.12.cu13" "$(run_select "$(printf '9.0\n10.3')")"
|
|
assert_eq "B200 then GB10 -> cu13 NVRTC selected" "UNSET libnvrtc.so.12.cu13" "$(run_select "$(printf '10.0\n12.1')")"
|
|
assert_eq "B300 then H100 -> cu13 NVRTC selected" "UNSET libnvrtc.so.12.cu13" "$(run_select "$(printf '10.3\n9.0')")"
|
|
assert_eq "H100 then A100 -> cu128 default kept" "UNSET libnvrtc.so.12.cu128.orig" "$(run_select "$(printf '9.0\n8.0')")"
|
|
|
|
# Stateful transition: a cu13 selection left in the same container's writable
|
|
# layer by an earlier sm_103/sm_121 boot must be reversed when the container
|
|
# later starts on an ordinary GPU (or none) -- a 570-579 driver cannot load
|
|
# cu13-produced cubins -- and kept when the datacenter Blackwell is still there.
|
|
assert_eq "A100 after B300 -> cu128 restored" "UNSET libnvrtc.so.12.cu128.orig" "$(run_select 8.0 libnvrtc.so.12.cu13)"
|
|
assert_eq "no GPU after B300 -> cu128 restored" "UNSET libnvrtc.so.12.cu128.orig" "$(run_select none libnvrtc.so.12.cu13)"
|
|
assert_eq "B300 after B300 -> cu13 kept" "UNSET libnvrtc.so.12.cu13" "$(run_select 10.3 libnvrtc.so.12.cu13)"
|
|
|
|
rm -f "$_FUNC_FILE"
|
|
|
|
echo ""
|
|
echo "Results: $PASS passed, $FAIL failed"
|
|
[ "$FAIL" -eq 0 ] || exit 1
|