From 46c713cdd3cec956ac991d2418186ec4a3e4bed0 Mon Sep 17 00:00:00 2001 From: Valeri Sitnikov Date: Mon, 21 Sep 2026 07:10:31 +0100 Subject: [PATCH] Regression runner on the server with pytest-xdist -n 14 (installed into ~/torchenv with Valeri's consent) and BLAS threads pinned to 1: 755 passed in 2:00 and 2:04, against 5:49 single-process on the server and 3:42 on the laptop. Single-process pytest on the server is slower than the laptop; the server only pays with parallelism. Co-Authored-By: Claude Opus 5 --- run_regression.sh | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/run_regression.sh b/run_regression.sh index ac9e3cc..843210d 100644 --- a/run_regression.sh +++ b/run_regression.sh @@ -1,9 +1,14 @@ #!/bin/bash # Frozen regression runner: detached, logged, with an explicit end marker, so an ssh # drop cannot kill it (2026-09-21: a foreground pytest over ssh died with the connection). +# +# -n 14 via pytest-xdist: single-process pytest on the server is SLOWER than the laptop +# (5:49 against 3:42), so the server only pays with parallelism. BLAS threads MUST be +# pinned to 1, or every xdist worker grabs 3+ cores (2026-09-16: 50+ minutes, load 55/16). +export OMP_NUM_THREADS=1 OPENBLAS_NUM_THREADS=1 MKL_NUM_THREADS=1 cd "$HOME/sim-numerics" || exit 1 for pass in 1 2; do echo "=== PASS $pass ===" - ~/torchenv/bin/python -m pytest tests -q -p no:cacheprovider 2>&1 | tail -4 + ~/torchenv/bin/python -m pytest tests -q -p no:cacheprovider -n 14 2>&1 | tail -4 done echo "=== REGRESSION DONE ==="