Skip to content

Commit de6d31f

Browse files
committed
AZP/RELEASE: Add CUDA 13 support
1 parent 1a43fda commit de6d31f

10 files changed

+88
-5
lines changed

buildlib/az-distro-release.yml

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -37,6 +37,12 @@ jobs:
3737
ubuntu24_cuda12_${{ parameters.arch }}:
3838
build_container: ubuntu24_cuda12_${{ parameters.arch }}
3939
artifact_name: $(POSTFIX)-ubuntu24.04-mofed5-cuda12-${{ parameters.arch }}.tar.bz2
40+
ubuntu22_cuda13_${{ parameters.arch }}:
41+
build_container: ubuntu22_cuda13_${{ parameters.arch }}
42+
artifact_name: $(POSTFIX)-ubuntu22.04-mofed5-cuda13-${{ parameters.arch }}.tar.bz2
43+
ubuntu24_cuda13_${{ parameters.arch }}:
44+
build_container: ubuntu24_cuda13_${{ parameters.arch }}
45+
artifact_name: $(POSTFIX)-ubuntu24.04-mofed5-cuda13-${{ parameters.arch }}.tar.bz2
4046
# x86 only
4147
${{ if eq(parameters.arch, 'x86_64') }}:
4248
centos7_cuda11_${{ parameters.arch }}:

buildlib/az-helpers.sh

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -189,13 +189,13 @@ try_load_cuda_env() {
189189
[ "${num_gpus}" -gt 0 ] || return 0
190190

191191
# Check cuda env module
192-
az_module_load dev/cuda12.8 || return 0
192+
az_module_load dev/cuda13.0.0 || return 0
193193
have_cuda=yes
194194

195195
# Check gdrcopy
196196
if [ -w "/dev/gdrdrv" ]
197197
then
198-
az_module_load dev/gdrcopy2.4.4_cuda12.8.0 && have_gdrcopy=yes
198+
az_module_load dev/gdrcopy2.5.1_cuda13.0.0 && have_gdrcopy=yes
199199
fi
200200
}
201201

buildlib/azure-pipelines-release-drp.yml

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -44,6 +44,10 @@ resources:
4444
image: $(REPO_MIRROR)/ucx/x86_64/ubuntu20.04-mofed5-cuda12:3
4545
- container: ubuntu24_cuda12_x86_64
4646
image: $(REPO_MIRROR)/ucx/x86_64/ubuntu24.04-mofed24.10-cuda12.5:1
47+
- container: ubuntu22_cuda13_x86_64
48+
image: $(REPO_MIRROR)/ucx/x86_64/ubuntu22.04-mofed5-cuda13:1
49+
- container: ubuntu24_cuda13_x86_64
50+
image: $(REPO_MIRROR)/ucx/x86_64/ubuntu24.04-mofed24.10-cuda13:1
4751

4852
# aarch64
4953
- container: centos8_cuda11_aarch64
@@ -61,6 +65,10 @@ resources:
6165
image: $(REPO_MIRROR)/ucx/aarch64/ubuntu22.04-mofed5-cuda12:3
6266
- container: ubuntu24_cuda12_aarch64
6367
image: $(REPO_MIRROR)/ucx/aarch64/ubuntu24.04-mofed24.10-cuda12.5:1
68+
- container: ubuntu22_cuda13_aarch64
69+
image: $(REPO_MIRROR)/ucx/aarch64/ubuntu22.04-mofed5-cuda13:1
70+
- container: ubuntu24_cuda13_aarch64
71+
image: $(REPO_MIRROR)/ucx/aarch64/ubuntu24.04-mofed24.10-cuda13:1
6472

6573
stages:
6674
- stage: Prepare

buildlib/azure-pipelines-release.yml

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -40,6 +40,10 @@ resources:
4040
image: rdmz-harbor.rdmz.labs.mlnx/ucx/x86_64/ubuntu22.04-mofed5-cuda12:3
4141
- container: ubuntu24_cuda12_x86_64
4242
image: rdmz-harbor.rdmz.labs.mlnx/ucx/x86_64/ubuntu24.04-mofed24.10-cuda12.5:1
43+
- container: ubuntu22_cuda13_x86_64
44+
image: rdmz-harbor.rdmz.labs.mlnx/ucx/x86_64/ubuntu22.04-mofed5-cuda13:1
45+
- container: ubuntu24_cuda13_x86_64
46+
image: rdmz-harbor.rdmz.labs.mlnx/ucx/x86_64/ubuntu24.04-mofed24.10-cuda13:1
4347

4448
# aarch64
4549
- container: centos8_cuda11_aarch64
@@ -57,6 +61,10 @@ resources:
5761
image: rdmz-harbor.rdmz.labs.mlnx/ucx/aarch64/ubuntu22.04-mofed5-cuda12:3
5862
- container: ubuntu24_cuda12_aarch64
5963
image: rdmz-harbor.rdmz.labs.mlnx/ucx/aarch64/ubuntu24.04-mofed24.10-cuda12.5:1
64+
- container: ubuntu22_cuda13_aarch64
65+
image: rdmz-harbor.rdmz.labs.mlnx/ucx/aarch64/ubuntu22.04-mofed5-cuda13:1
66+
- container: ubuntu24_cuda13_aarch64
67+
image: rdmz-harbor.rdmz.labs.mlnx/ucx/aarch64/ubuntu24.04-mofed24.10-cuda13:1
6068

6169
stages:
6270
- stage: Prepare

buildlib/dockers/docker-compose-aarch64.yml

Lines changed: 25 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -93,3 +93,28 @@ services:
9393
CUDA_VERSION: 12.5.1
9494
NV_DRIVER_VERSION: 555
9595
ARCH: aarch64
96+
97+
ubuntu22.04-mofed5-cuda13:
98+
image: ubuntu22.04-mofed5-cuda13:1
99+
build:
100+
context: .
101+
network: host
102+
dockerfile: ubuntu-release.Dockerfile
103+
args:
104+
MOFED_VERSION: 5.8-7.0.6.1
105+
UBUNTU_VERSION: 22.04
106+
CUDA_VERSION: 13.0.0
107+
NV_DRIVER_VERSION: 580
108+
ARCH: aarch64
109+
ubuntu24.04-mofed5-cuda13:
110+
image: ubuntu24.04-mofed24.10-cuda13:1
111+
build:
112+
context: .
113+
network: host
114+
dockerfile: ubuntu-release.Dockerfile
115+
args:
116+
MOFED_VERSION: 24.10-3.2.5.0
117+
UBUNTU_VERSION: 24.04
118+
CUDA_VERSION: 13.0.0
119+
NV_DRIVER_VERSION: 580
120+
ARCH: aarch64

buildlib/dockers/docker-compose-x86_64.yml

Lines changed: 26 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -129,3 +129,29 @@ services:
129129
CUDA_VERSION: 12.5.1
130130
NV_DRIVER_VERSION: 555
131131
ARCH: x86_64
132+
133+
134+
ubuntu22.04-mofed5-cuda13:
135+
image: ubuntu22.04-mofed5-cuda13:1
136+
build:
137+
context: .
138+
network: host
139+
dockerfile: ubuntu-release.Dockerfile
140+
args:
141+
MOFED_VERSION: 5.8-7.0.6.1
142+
UBUNTU_VERSION: 22.04
143+
CUDA_VERSION: 13.0.0
144+
NV_DRIVER_VERSION: 580
145+
ARCH: x86_64
146+
ubuntu24.04-mofed5-cuda13:
147+
image: ubuntu24.04-mofed24.10-cuda13:1
148+
build:
149+
context: .
150+
network: host
151+
dockerfile: ubuntu-release.Dockerfile
152+
args:
153+
MOFED_VERSION: 24.10-3.2.5.0
154+
UBUNTU_VERSION: 24.04
155+
CUDA_VERSION: 13.0.0
156+
NV_DRIVER_VERSION: 580
157+
ARCH: x86_64

buildlib/pr/cuda/cuda.yml

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -92,6 +92,10 @@ jobs:
9292
CONTAINER: ubuntu22_cuda_12_0
9393
ubuntu22_cuda_12_1:
9494
CONTAINER: ubuntu22_cuda_12_1
95+
ubuntu22_cuda_13_0:
96+
CONTAINER: ubuntu22_cuda_13_0
97+
ubuntu24_cuda_13_0:
98+
CONTAINER: ubuntu24_cuda_13_0
9599

96100
container: $[ variables['CONTAINER'] ]
97101
timeoutInMinutes: 35

buildlib/pr/main.yml

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -174,6 +174,12 @@ resources:
174174
- container: ubuntu22_cuda12
175175
image: rdmz-harbor.rdmz.labs.mlnx/ucx/x86_64/ubuntu22.04-mofed5-cuda12:3
176176
options: $(DOCKER_OPT_ARGS) $(DOCKER_OPT_VOLUMES) $(DOCKER_OPT_GPU)
177+
- container: ubuntu22_cuda13
178+
image: rdmz-harbor.rdmz.labs.mlnx/ucx/x86_64/ubuntu22.04-mofed5-cuda13:1
179+
options: $(DOCKER_OPT_ARGS) $(DOCKER_OPT_VOLUMES) $(DOCKER_OPT_GPU)
180+
- container: ubuntu24_cuda13
181+
image: rdmz-harbor.rdmz.labs.mlnx/ucx/x86_64/ubuntu24.04-mofed24.10-cuda13:1
182+
options: $(DOCKER_OPT_ARGS) $(DOCKER_OPT_VOLUMES) $(DOCKER_OPT_GPU)
177183
- container: ubuntu2204_rocm_6_0_0
178184
image: rdmz-harbor.rdmz.labs.mlnx/ucx/x86_64/ubuntu2204:rocm-6.0.0
179185
options: $(DOCKER_OPT_ARGS) $(DOCKER_OPT_VOLUMES)

buildlib/tools/common.sh

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -4,8 +4,8 @@ WORKSPACE=${WORKSPACE:=$PWD}
44
# build in local directory which goes away when docker exits
55
ucx_build_dir=$HOME/${BUILD_ID}/build
66
ucx_inst=$ucx_build_dir/install
7-
CUDA_MODULE="dev/cuda12.8"
8-
GDRCOPY_MODULE="dev/gdrcopy2.4.4_cuda12.8.0"
7+
CUDA_MODULE="dev/cuda13.0.0"
8+
GDRCOPY_MODULE="dev/gdrcopy2.5.1_cuda13.0.0"
99
JDK_MODULE="dev/jdk"
1010
MVN_MODULE="dev/mvn"
1111
XPMEM_MODULE="dev/xpmem-90a95a4"

buildlib/tools/perf-common.yml

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -13,7 +13,7 @@ steps:
1313
1414
case "${{ parameters.Name }}" in
1515
"Build-UCX")
16-
module="/hpc/local/etc/modulefiles/dev/cuda12.8"
16+
module="/hpc/local/etc/modulefiles/dev/cuda13.0.0"
1717
perfxParams=(--skip-run --source-branch $(Build.SourceBranch) --omb-cuda)
1818
;;
1919
"Perf-test-multi-node")

0 commit comments

Comments
 (0)