| 24 Mar 2025 |
 Moritz Sanft | Are you sure that removing the OpenSSL1.1 compatibility libraries from the NVIDIA driver package is correct?
see: https://github.com/NixOS/nixpkgs/blob/a13cc0bc42a89a025f545c4a17f25e6300710f39/pkgs/os-specific/linux/nvidia-x11/builder.sh#L81-L82
I ran into PyTorch 2.6 trying to open that library today, hence the question. | 15:41:42 |
| Moritz Sanft | * Are we sure that removing the OpenSSL1.1 compatibility libraries from the NVIDIA driver package is correct?
see: https://github.com/NixOS/nixpkgs/blob/a13cc0bc42a89a025f545c4a17f25e6300710f39/pkgs/os-specific/linux/nvidia-x11/builder.sh#L81-L82
I ran into PyTorch 2.6 trying to open that library today, hence the question. | 15:41:49 |
 SomeoneSerge (back on matrix) | Hm. The user would have to manually enable patchelfing the paths to openssl 1.1 anyway, I'd say we shouldn't include them y default because then everything goes behind permittedInsecurePackages or what's the name | 15:43:20 |
| SomeoneSerge (back on matrix) |
I ran into PyTorch 2.6 trying to open that library today, hence the question.
Can you also file a report with the logs on github?
| 15:43:58 |
| Moritz Sanft | Sure! | 15:47:58 |
 Gaétan Lepage | Hi,
I'm looking at mistral-rs and trying to make autoAddDriverRunPath and autoPatchelfHook work, but I get this at runtime:
Unable to dynamically load the "cuda" shared library - searched for library names: ["cuda", "nvcuda"]. Ensure that `LD_LIBRARY_PATH` has the correct path to the installed library. If the shared library is present on the system under a different name than one of those listed above, please open a GitHub issue.
| 16:30:22 |
| SomeoneSerge (back on matrix) | They probably CDLL on the python side? | 17:26:26 |
| Gaétan Lepage | It's rust | 17:43:35 |
| Gaétan Lepage | But yeah, they surely DLopen | 17:43:40 |
| 25 Mar 2025 |
 connor (he/him) | Kevin Mittman: I really like that cuda_cudart provides stubs for driver libraries unavailable at build time — any chance we could see stubs for core Jetson libraries like libnvdla_compiler.so or libnvdla_runtime.so, or whether something like llvm-ifs (https://llvm.org/docs/CommandGuide/llvm-ifs.html) would be able to generate stubs from unpacked shared object files from the Jetson Debian packages? | 06:55:39 |
| SomeoneSerge (back on matrix) |
We’d also like to extend an open invitation for anyone who wishes to help us maintain Xen!
Let us know if you’d like to help us test, update and keep Xen working for the foreseeable future! Simply open a Pull Request on Nixpkgs adding yourself to the Xen team on maintainers/team-list.nix.
Oh, that's how it's done. I don't believe we ever did this CC connor (he/him) (UTC-7)
| 11:17:38 |
| Gaétan Lepage | Redacted or Malformed Event | 12:49:38 |
| Gaétan Lepage | Redacted or Malformed Event | 12:49:39 |
| Gaétan Lepage | Redacted or Malformed Event | 12:51:52 |
| Gaétan Lepage | Redacted or Malformed Event | 12:51:55 |
| 26 Mar 2025 |
|  kaya 𖤐 changed their profile picture. | 15:33:34 |
| 28 Mar 2025 |
 stick | 
Download Captura de pantalla 2025-03-28 a la(s) 09.50.22.png | 08:50:45 |
| stick | we now have CUDA 12.8 in master and number of succeeding jobs is over 1000+ 🎉 | 08:50:46 |
| Gaétan Lepage | Well done stick!! | 08:52:31 |
 mcwitt | I'm running into some trouble linking programs that use dynamic parallelism (CDP) using the Nix CUDA toolchain: the linker fails to find a couple of CDP symbols, e.g.
nvlink error : Undefined reference to '__cudaCDP2GetParameterBufferV2' in '/tmp/tmpxft_002f57e3_00000000-11_test_cdp.o'
nvlink error : Undefined reference to '__cudaCDP2LaunchDeviceV2' in '/tmp/tmpxft_002f57e3_00000000-11_test_cdp.o'
I put together a minimal repro here: https://gist.github.com/mcwitt/94b13848ffdf762c587e9bea1257c02b
I see some issues have been mentioned in the past, but no resolution was noted: https://github.com/NVIDIA/cuda-samples/issues/264
| 21:39:15 |
| mcwitt | * I'm running into some trouble linking programs that use dynamic parallelism (CDP) using the NixOS CUDA toolchain: the linker fails to find a couple of CDP symbols, e.g.
nvlink error : Undefined reference to '__cudaCDP2GetParameterBufferV2' in '/tmp/tmpxft_002f57e3_00000000-11_test_cdp.o'
nvlink error : Undefined reference to '__cudaCDP2LaunchDeviceV2' in '/tmp/tmpxft_002f57e3_00000000-11_test_cdp.o'
I put together a minimal repro here: https://gist.github.com/mcwitt/94b13848ffdf762c587e9bea1257c02b
I see some issues have been mentioned in the past, but no resolution was noted: https://github.com/NVIDIA/cuda-samples/issues/264
| 21:40:49 |
|  eugeniusz joined the room. | 23:33:31 |
| 29 Mar 2025 |
 ruro | I think I saw very similar error messages while trying to fix cuda-samples and I was able to fix them by explicitly adding a -L${CUDA_PATH}/lib to the nvcc invocation, where CUDA_PATH is a symlinkJoin with a bunch of cuda libs (not 100% sure, which specific ones were needed for the __cudaCDP2* symbols, maybe cuda_cudart or cuda_cudart.static?). | 07:26:15 |
| 1 Apr 2025 |
|  Artem Andreenko joined the room. | 09:58:12 |
|  Adam Neverwas set a profile picture. | 23:15:44 |
| 2 Apr 2025 |
|  @athyfr:matrix.org joined the room. | 01:29:37 |
| 3 Apr 2025 |
|  theabm joined the room. | 12:00:11 |
| 5 Apr 2025 |
|  underpantsgnome left the room. | 15:47:02 |
| 8 Apr 2025 |
| SomeoneSerge (back on matrix) | Kevin Mittman: are you aware of https://github.com/systemd/systemd/pull/32234/? It woudl be great to have this for CUDAToolkit and other components | 14:25:50 |
| connor (he/him) | Hey all, some notes from a call I had with Serge about some priorities we're tracking: https://pad.lassul.us/s/GBcXdgeFz | 16:15:57 |
