{system}[system/system] CUDA v13.3.0

[Thyre]

(created using eb --new-pr)

This CUDA version supports LLVM 21.x and GCC 15. So this will likely be the CUDA version used for 2026.1.
CUDA Tile C++ is also certainly interesting. As a last point, this fixes the regressions seen in CUDA 13.1 and 13.2 (see here)

[Thyre]

@boegelbot please test @ jsc-zen3

[boegelbot]

@Thyre: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de

PR test command 'if [[ develop != 'develop' ]]; then EB_BRANCH=develop ./easybuild_develop.sh 2> /dev/null 1>&2; EB_PREFIX=/home/boegelbot/easybuild/develop source init_env_easybuild_develop.sh; fi; EB_PR=26090 EB_ARGS= EB_CONTAINER= EB_REPO=easybuild-easyconfigs EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_26090 --ntasks=8 ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

• exit code: 0
• output:

Submitted batch job 10505

Test results coming soon (I hope)...

Details

- notification for comment with ID 4552204153 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

[github-actions]

Updated software CUDA-13.3.0.eb

Diff against CUDA-13.2.0.eb

easybuild/easyconfigs/c/CUDA/CUDA-13.2.0.eb

diff --git a/easybuild/easyconfigs/c/CUDA/CUDA-13.2.0.eb b/easybuild/easyconfigs/c/CUDA/CUDA-13.3.0.eb
index f8a66c40fd..76a2f62c8c 100644
--- a/easybuild/easyconfigs/c/CUDA/CUDA-13.2.0.eb
+++ b/easybuild/easyconfigs/c/CUDA/CUDA-13.3.0.eb
@@ -1,6 +1,6 @@
 name = 'CUDA'
-version = '13.2.0'
-local_nv_version = '595.45.04'
+version = '13.3.0'
+local_nv_version = '610.43.02'
 
 homepage = 'https://developer.nvidia.com/cuda-toolkit'
 description = """CUDA (formerly Compute Unified Device Architecture) is a parallel
@@ -14,9 +14,9 @@ source_urls = ['https://developer.download.nvidia.com/compute/cuda/%(version)s/l
 sources = ['cuda_%%(version)s_%s_linux%%(cudaarch)s.run' % local_nv_version]
 checksums = [{
     'cuda_%%(version)s_%s_linux.run' % local_nv_version:
-        'c599b651582c8f345ebde7bcf089b7dcfcd8afd7d3ea7f50525698c563a239d7',
+        '5f79488b57fe6936bc95a56f9b7e2838ab2f2ee3313b1008942206eebe06352d',
     'cuda_%%(version)s_%s_linux_sbsa.run' % local_nv_version:
-        '9684ebc64988f71ef1c70846cc9d158fb27950c1355bf37c8795c346346b6a72',
+        '94ec4572197b65532dcf3d327460417c6527fa42ded9d5010e06ddb89e878d4c'
 }]
 
 moduleclass = 'system'

Diff against CUDA-13.1.1.eb

easybuild/easyconfigs/c/CUDA/CUDA-13.1.1.eb

diff --git a/easybuild/easyconfigs/c/CUDA/CUDA-13.1.1.eb b/easybuild/easyconfigs/c/CUDA/CUDA-13.3.0.eb
index 725d546dcc..76a2f62c8c 100644
--- a/easybuild/easyconfigs/c/CUDA/CUDA-13.1.1.eb
+++ b/easybuild/easyconfigs/c/CUDA/CUDA-13.3.0.eb
@@ -1,6 +1,6 @@
 name = 'CUDA'
-version = '13.1.1'
-local_nv_version = '590.48.01'
+version = '13.3.0'
+local_nv_version = '610.43.02'
 
 homepage = 'https://developer.nvidia.com/cuda-toolkit'
 description = """CUDA (formerly Compute Unified Device Architecture) is a parallel
@@ -14,9 +14,9 @@ source_urls = ['https://developer.download.nvidia.com/compute/cuda/%(version)s/l
 sources = ['cuda_%%(version)s_%s_linux%%(cudaarch)s.run' % local_nv_version]
 checksums = [{
     'cuda_%%(version)s_%s_linux.run' % local_nv_version:
-        '24ff323723722781436804b392a48f691cb40de9808095d3e2192d0db6dfb8e4',
+        '5f79488b57fe6936bc95a56f9b7e2838ab2f2ee3313b1008942206eebe06352d',
     'cuda_%%(version)s_%s_linux_sbsa.run' % local_nv_version:
-        '8adcd5d4b3e1e70f7420959b97514c0c97ec729da248d54902174c4d229bfd2c'
+        '94ec4572197b65532dcf3d327460417c6527fa42ded9d5010e06ddb89e878d4c'
 }]
 
 moduleclass = 'system'

Diff against CUDA-13.1.0.eb

easybuild/easyconfigs/c/CUDA/CUDA-13.1.0.eb

diff --git a/easybuild/easyconfigs/c/CUDA/CUDA-13.1.0.eb b/easybuild/easyconfigs/c/CUDA/CUDA-13.3.0.eb
index a276ae305f..76a2f62c8c 100644
--- a/easybuild/easyconfigs/c/CUDA/CUDA-13.1.0.eb
+++ b/easybuild/easyconfigs/c/CUDA/CUDA-13.3.0.eb
@@ -1,6 +1,6 @@
 name = 'CUDA'
-version = '13.1.0'
-local_nv_version = '590.44.01'
+version = '13.3.0'
+local_nv_version = '610.43.02'
 
 homepage = 'https://developer.nvidia.com/cuda-toolkit'
 description = """CUDA (formerly Compute Unified Device Architecture) is a parallel
@@ -14,9 +14,9 @@ source_urls = ['https://developer.download.nvidia.com/compute/cuda/%(version)s/l
 sources = ['cuda_%%(version)s_%s_linux%%(cudaarch)s.run' % local_nv_version]
 checksums = [{
     'cuda_%%(version)s_%s_linux.run' % local_nv_version:
-        '6b4fdf2694b3d7afbc526f26412b4cf4f050b202324455053307310f53b323a7',
+        '5f79488b57fe6936bc95a56f9b7e2838ab2f2ee3313b1008942206eebe06352d',
     'cuda_%%(version)s_%s_linux_sbsa.run' % local_nv_version:
-        '06cda49a7031b1c99f784237be5c852619379cbba9555036045044b9ddc99240',
+        '94ec4572197b65532dcf3d327460417c6527fa42ded9d5010e06ddb89e878d4c'
 }]
 
 moduleclass = 'system'

[Thyre]

Test report by @Thyre
SUCCESS
Build succeeded for 1 out of 1 (total: 1 min 55 secs) (1 easyconfigs in total)
jpbl-s03-02 - Linux RHEL 9.7, AArch64, ARM UNKNOWN (neoverse_v2), 1 x NVIDIA NVIDIA GH200 480GB, 595.71.05, Python 3.9.25
See https://gist.github.com/Thyre/3e810d5a4bff1f8c807e8c5d35156c35 for a full test report.

[boegelbot]

Test report by @boegelbot
SUCCESS
Build succeeded for 1 out of 1 (total: 8 mins 35 secs) (1 easyconfigs in total)
jsczen3c1.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.7, x86_64, AMD EPYC-Milan Processor (zen3), Python 3.9.25
See https://gist.github.com/boegelbot/25379f0cb463471a9ba9a50093630f41 for a full test report.

[Micket]

lgtm

[Micket]

Test report by @Micket
SUCCESS
Build succeeded for 1 out of 1 (total: 4 mins 39 secs) (1 easyconfigs in total)
vera-icelake-build - Linux Rocky Linux 9.6, x86_64, Intel(R) Xeon(R) Silver 4316 CPU @ 2.30GHz, Python 3.9.21
See https://gist.github.com/Micket/6f5cdc79cc865c20f134b10a2aa0dd4f for a full test report.

[Micket]

Going in, thanks @Thyre!