Advance docker builds to cuda 11.8 (#104528)

Advance docker builds to cuda 11.8
This should fix Docker build nightly failure: https://hud.pytorch.org/hud/pytorch/pytorch/nightly/1?per_page=50&name_filter=Docker

The following docker image no longer exists: ``nvidia/cuda:11.7.0-cudnn8-devel-ubuntu20.04``
Hence advancing build to ``nvidia/cuda/11.8.0-cudnn8-devel-ubuntu20.04``
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104528
Approved by: https://github.com/DanilBaibak
This commit is contained in:
atalman
2023-07-03 16:44:23 +00:00
committed by PyTorch MergeBot
parent d6b1f12846
commit 0ac2666d72

View File

@ -8,7 +8,7 @@ $(warning WARNING: No docker user found using results from whoami)
DOCKER_ORG = $(shell whoami)
endif
CUDA_VERSION = 11.7.0
CUDA_VERSION = 11.8.0
CUDNN_VERSION = 8
BASE_RUNTIME = ubuntu:20.04
BASE_DEVEL = nvidia/cuda:$(CUDA_VERSION)-cudnn$(CUDNN_VERSION)-devel-ubuntu20.04