Solution to issue cannot be found in the documentation.
Issue
The conda-forge versions of triton 3.2 is unusable for me. I end up with a segmentation fault when compiling from TTGIR to LLIR. Triton 3.1 did not cause any issues, whereas triton 3.3, the LLIR compilation step comes with a "Failures have been detected while processing an MLIR pass pipeline" error message. However, this might be unrelated. I tried to set things up with python 3.11, 3.12 and 3.13, but without success (note that I didn't actually try all possible combinations of triton and python versions).
After re-installing triton with pip (pip uninstall triton && pip install triton==3.2), the issue with the segmentation fault is resolved. Therefore, I suspect that this is a packaging issue. I tried something similar for triton 3.3, but this did not resolve the issue there, which seems to suggest that my triton 3.3 issues are not (directly) related to the segmentation fault.
PS: Triton also depends on cuda-cudart-dev on systems without a system-wide cuda install (for cuda.h). Should I create a separate issue for this?
Installed packages
_openmp_mutex 4.5 3_kmp_llvm conda-forge
asttokens 3.0.0 pyhd8ed1ab_1 conda-forge
attr 2.5.1 h166bdaf_1 conda-forge
bzip2 1.0.8 h4bc722e_7 conda-forge
ca-certificates 2025.8.3 hbd8a1cb_0 conda-forge
cpython 3.11.13 py311hd8ed1ab_0 conda-forge
cuda-cccl_linux-64 12.9.27 ha770c72_0 conda-forge
cuda-crt-tools 12.9.86 ha770c72_2 conda-forge
cuda-cudart 12.9.79 h5888daf_0 conda-forge
cuda-cudart-dev 12.9.79 h5888daf_0 conda-forge
cuda-cudart-dev_linux-64 12.9.79 h3f2d84a_0 conda-forge
cuda-cudart-static 12.9.79 h5888daf_0 conda-forge
cuda-cudart-static_linux-64 12.9.79 h3f2d84a_0 conda-forge
cuda-cudart_linux-64 12.9.79 h3f2d84a_0 conda-forge
cuda-cuobjdump 12.9.82 hbd13f7d_0 conda-forge
cuda-cupti 12.9.79 h9ab20c4_0 conda-forge
cuda-nvcc-tools 12.9.86 he02047a_2 conda-forge
cuda-nvdisasm 12.9.88 hbd13f7d_0 conda-forge
cuda-nvrtc 12.9.86 h5888daf_0 conda-forge
cuda-nvtx 12.9.79 h5888daf_0 conda-forge
cuda-nvvm-tools 12.9.86 h4bc722e_2 conda-forge
cuda-version 12.9 h4f385c5_3 conda-forge
cudnn 9.11.0.98 hbcb9cd8_0 conda-forge
decorator 5.2.1 pyhd8ed1ab_0 conda-forge
exceptiongroup 1.3.0 pyhd8ed1ab_0 conda-forge
executing 2.2.0 pyhd8ed1ab_0 conda-forge
filelock 3.18.0 pyhd8ed1ab_0 conda-forge
fsspec 2025.7.0 pyhd8ed1ab_0 conda-forge
gmp 6.3.0 hac33072_2 conda-forge
gmpy2 2.2.1 py311h0f6cedb_0 conda-forge
ipython 9.4.0 pyhfa0c392_0 conda-forge
ipython_pygments_lexers 1.1.1 pyhd8ed1ab_0 conda-forge
jedi 0.19.2 pyhd8ed1ab_1 conda-forge
jinja2 3.1.6 pyhd8ed1ab_0 conda-forge
ld_impl_linux-64 2.44 h1423503_1 conda-forge
libabseil 20250127.1 cxx17_hbbce691_0 conda-forge
libblas 3.9.0 33_hfdb39a5_mkl conda-forge
libcap 2.75 h39aace5_0 conda-forge
libcblas 3.9.0 33_h372d94f_mkl conda-forge
libcublas 12.9.1.4 h9ab20c4_0 conda-forge
libcudnn 9.11.0.98 hf7e9902_0 conda-forge
libcudnn-dev 9.11.0.98 h58dd1b1_0 conda-forge
libcudss 0.5.0.16 h14340ca_1 conda-forge
libcufft 11.4.1.4 h5888daf_0 conda-forge
libcufile 1.14.1.1 ha8da6e3_0 conda-forge
libcurand 10.3.10.19 h9ab20c4_0 conda-forge
libcusolver 11.7.5.82 h9ab20c4_1 conda-forge
libcusparse 12.5.10.65 h5888daf_1 conda-forge
libexpat 2.7.1 hecca717_0 conda-forge
libffi 3.4.6 h2dba641_1 conda-forge
libgcc 15.1.0 h767d61c_4 conda-forge
libgcc-ng 15.1.0 h69a702a_4 conda-forge
libgcrypt-lib 1.11.1 hb9d3cd8_0 conda-forge
libgpg-error 1.55 h3f2d84a_0 conda-forge
libhwloc 2.12.1 default_h3d81e11_1000 conda-forge
libiconv 1.18 h4ce23a2_1 conda-forge
liblapack 3.9.0 33_hc41d3b0_mkl conda-forge
libllvm20 20.1.8 hecd9e04_0 conda-forge
liblzma 5.8.1 hb9d3cd8_2 conda-forge
libmagma 2.9.0 h9918c94_2 conda-forge
libnl 3.11.0 hb9d3cd8_0 conda-forge
libnsl 2.0.1 hb9d3cd8_1 conda-forge
libnvjitlink 12.9.86 h5888daf_1 conda-forge
libprotobuf 5.29.3 h7460b1f_2 conda-forge
libsqlite 3.50.4 h0c1763c_0 conda-forge
libstdcxx 15.1.0 h8f9b012_4 conda-forge
libstdcxx-ng 15.1.0 h4852527_4 conda-forge
libsystemd0 257.7 h4e0b6ca_0 conda-forge
libtorch 2.6.0 cuda126_mkl_h99b69db_304 conda-forge
libudev1 257.7 hbe16f8c_0 conda-forge
libuuid 2.38.1 h0b41bf4_0 conda-forge
libuv 1.51.0 hb03c661_1 conda-forge
libxcrypt 4.4.36 hd590300_1 conda-forge
libxml2 2.13.8 h2cb61b6_1 conda-forge
libzlib 1.3.1 hb9d3cd8_2 conda-forge
llvm-openmp 20.1.8 h4922eb0_1 conda-forge
lz4-c 1.10.0 h5888daf_1 conda-forge
markupsafe 3.0.2 py311h2dc5d0c_1 conda-forge
matplotlib-inline 0.1.7 pyhd8ed1ab_1 conda-forge
mkl 2024.2.2 ha770c72_17 conda-forge
mpc 1.3.1 h24ddda3_1 conda-forge
mpfr 4.2.1 h90cbb55_3 conda-forge
mpmath 1.3.0 pyhd8ed1ab_1 conda-forge
nccl 2.27.7.1 h49b9d9a_0 conda-forge
ncurses 6.5 h2d0b736_3 conda-forge
networkx 3.5 pyhe01879c_0 conda-forge
ninja 1.13.1 h171cf75_0 conda-forge
numpy 2.3.2 py311h2e04523_0 conda-forge
openssl 3.5.2 h26f9b46_0 conda-forge
optree 0.17.0 py311hdf67eae_0 conda-forge
parso 0.8.4 pyhd8ed1ab_1 conda-forge
pexpect 4.9.0 pyhd8ed1ab_1 conda-forge
pickleshare 0.7.5 pyhd8ed1ab_1004 conda-forge
pip 25.2 pyh8b19718_0 conda-forge
prompt-toolkit 3.0.51 pyha770c72_0 conda-forge
ptyprocess 0.7.0 pyhd8ed1ab_1 conda-forge
pure_eval 0.2.3 pyhd8ed1ab_1 conda-forge
pybind11 3.0.0 pyh9380348_1 conda-forge
pybind11-global 3.0.0 pyhf748d72_1 conda-forge
pygments 2.19.2 pyhd8ed1ab_0 conda-forge
python 3.11.13 h9e4cc4f_0_cpython conda-forge
python_abi 3.11 8_cp311 conda-forge
pytorch 2.6.0 cuda126_mkl_py311_hcada2b2_304 conda-forge
pytorch-gpu 2.6.0 cuda126_mkl_ha999a5f_304 conda-forge
rdma-core 58.0 h5888daf_0 conda-forge
readline 8.2 h8c095d6_2 conda-forge
setuptools 75.8.2 pyhff2d567_0 conda-forge
sleef 3.8 h1b44611_0 conda-forge
stack_data 0.6.3 pyhd8ed1ab_1 conda-forge
sympy 1.14.0 pyh2585a3b_105 conda-forge
tbb 2021.13.0 hb60516a_2 conda-forge
tk 8.6.13 noxft_hd72426e_102 conda-forge
traitlets 5.14.3 pyhd8ed1ab_1 conda-forge
triton 3.2.0 cuda126py311hc9dd8b4_1 conda-forge
typing-extensions 4.14.1 h4440ef1_0 conda-forge
typing_extensions 4.14.1 pyhe01879c_0 conda-forge
tzdata 2025b h78e105d_0 conda-forge
wcwidth 0.2.13 pyhd8ed1ab_1 conda-forge
wheel 0.45.1 pyhd8ed1ab_1 conda-forge
zstd 1.5.7 hb8e6e7a_2 conda-forge
Environment info
active environment : pytorch26
active env location : ~/.conda/envs/pytorch26
shell level : 2
user config file : ~/.condarc
populated config files : ~/.condarc
conda version : 25.7.0
conda-build version : not installed
python version : 3.12.4.final.0
solver : libmamba (default)
virtual packages : __archspec=1=skylake
__conda=25.7.0=0
__cuda=12.9=0
__glibc=2.42=0
__linux=6.15.8=0
__unix=0=0
base environment : /opt/miniconda3 (writable)
conda av data dir : /opt/miniconda3/etc/conda
conda av metadata url : None
channel URLs : https://conda.anaconda.org/conda-forge/linux-64
https://conda.anaconda.org/conda-forge/noarch
package cache : ~/.conda/pkgs
/opt/miniconda3/pkgs
envs directories : ~/.conda/envs
/opt/miniconda3/envs
~/.conda/envs
platform : linux-64
user-agent : conda/25.7.0 requests/2.32.4 CPython/3.12.4 Linux/6.15.8-arch1-2 arch/ glibc/2.42 solver/libmamba conda-libmamba-solver/25.4.0 libmambapy/2.0.5
UID:GID : 1000:1000
netrc file : ~/.netrc
offline mode : False
Solution to issue cannot be found in the documentation.
Issue
The conda-forge versions of triton 3.2 is unusable for me. I end up with a segmentation fault when compiling from TTGIR to LLIR. Triton 3.1 did not cause any issues, whereas triton 3.3, the LLIR compilation step comes with a "Failures have been detected while processing an MLIR pass pipeline" error message. However, this might be unrelated. I tried to set things up with python 3.11, 3.12 and 3.13, but without success (note that I didn't actually try all possible combinations of triton and python versions).
After re-installing triton with pip (
pip uninstall triton && pip install triton==3.2), the issue with the segmentation fault is resolved. Therefore, I suspect that this is a packaging issue. I tried something similar for triton 3.3, but this did not resolve the issue there, which seems to suggest that my triton 3.3 issues are not (directly) related to the segmentation fault.PS: Triton also depends on
cuda-cudart-devon systems without a system-wide cuda install (forcuda.h). Should I create a separate issue for this?Installed packages
Environment info