Rebuild LightGBM for cuda sanity check

[casparvl]

No description provided.

[casparvl]

bot: build repo:eessi.io-2023.06-software instance:eessi-bot-surf for:arch=x86_64/intel/icelake,accel=nvidia/cc80

[eessi-bot-surf]

New job on instance eessi-bot-surf for repository eessi.io-2023.06-software
Building on: intel-icelake and accelerator nvidia/cc80
Building for: x86_64/intel/icelake and accelerator nvidia/cc80
Job dir: /projects/eessibot/eessi-bot-surf/jobs/2025.08/pr_1169/14349897

date	job status	comment
Aug 28 15:05:56 UTC 2025	submitted	job id 14349897 will be eligible to start in about 20 seconds
Aug 28 15:06:19 UTC 2025	received	job awaits launch by Slurm scheduler
Aug 28 15:06:26 UTC 2025	running	job 14349897 is running
Aug 28 15:40:29 UTC 2025	finished	😢 FAILURE (click triangle for details) Details ✅ job output file slurm-14349897.out ✅ no message matching FATAL: ❌ found message matching ERROR: ❌ found message matching FAILED: ❌ found message matching required modules missing: ❌ no message matching No missing installations ✅ found message matching .tar.gz created! Artefacts eessi-2023.06-software-linux-x86_64-intel-icelake-accel-nvidia-cc80-17563939620.tar.gz size: 0 MiB (45 bytes) entries: 0 modules under 2023.06/software/linux/x86_64/intel/icelake/accel/nvidia/cc80/modules/all no module files in tarball software under 2023.06/software/linux/x86_64/intel/icelake/accel/nvidia/cc80/software no software packages in tarball reprod directories under 2023.06/software/linux/x86_64/intel/icelake/accel/nvidia/cc80/reprod no reprod directories in tarball other under 2023.06/software/linux/x86_64/intel/icelake/accel/nvidia/cc80 no other files in tarball
Aug 28 15:40:29 UTC 2025	test result	😁 SUCCESS (click triangle for details) ReFrame Summary [ SKIP ] (1/8) Skipping GPU test : only 1 GPU available for this test case [ SKIP ] (2/8) Skipping GPU test : only 1 GPU available for this test case [ SKIP ] (3/8) Skipping GPU test : only 1 GPU available for this test case [ SKIP ] (4/8) Skipping GPU test : only 1 GPU available for this test case [ SKIP ] (5/8) Skipping test : 1 GPU(s) available for this test case, need exactly 2 [ SKIP ] (6/8) Skipping test : 1 GPU(s) available for this test case, need exactly 2 [ SKIP ] (7/8) Skipping test : 1 GPU(s) available for this test case, need exactly 2 [ SKIP ] (8/8) Skipping test : 1 GPU(s) available for this test case, need exactly 2 [ PASSED ] Ran 0/8 test case(s) from 8 check(s) (0 failure(s), 8 skipped, 0 aborted) Details ✅ job output file slurm-14349897.out ❌ found message matching ERROR: ✅ no message matching [\s*FAILED\s*].*Ran .* test case

[casparvl]

1 files missing one or more CUDA compute capabilities:
  lib/python3.11/site-packages/lightgbm/lib/lib_lightgbm.so
1 files with device code for more CUDA Compute Capabilities than requested:
  lib/python3.11/site-packages/lightgbm/lib/lib_lightgbm.so
1 files missing PTX code for the highest configured CUDA Compute Capability:
  lib/python3.11/site-packages/lightgbm/lib/lib_lightgbm.so

And:

Fatbin elf code:
================
arch = sm_60
code version = [1,7]
host = linux
compile_size = 64bit
compressed

[casparvl]

Also here, we need to figure out how to convince the build system to build for a different CUDA arch...

[bedroge]

This part in the CMakeLists.txt is probably relevant and needs to be patched: https://github.com/microsoft/LightGBM/blob/v4.5.0/CMakeLists.txt#L224

[ocaisa]

Replaced by #1205