Compile on blackwell with CUDA 13+ - #787
Closed
atvasilopoulos wants to merge 3 commits into
Closed
atvasilopoulos wants to merge 3 commits into
atvasilopoulos wants to merge 3 commits into
Conversation
Collaborator
|
Hello @atvasilopoulos ! Thanks for the PR, I think we are aiming for the same, can you check/review my previous PR addressing this problem, that will help a lot! #783 |
Collaborator
Author
|
I will close this since #783 was merged with a more complete solution to the problem. Thanks @PabloCarmona! |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Compiling from source did not work properly when trying to compile on a Blackwell GPU with CUDA 13+. Found that cub::TransformInputIterator and cub::CountingInputIterator are getting deprecated and are moving to the new calls cuda::transform_iterator and cuda::counting_iterator. Changed them appropriately with
CUDART_VERSION >= 13000guards.Details
Compiling on a Blackwell GPU using CUDA 13.2 failed due to errors with the cub iterators. Changed them to the new iterators in the CUDA namespace, and the repo compiles properly. Tests are passing.
When running tests found two warnings regarding the RangeEstimator class, which are now fixed (detaching before conversion to float).
GPU model:
RTX PRO 6000 Blackwell Workstation Edition
CUDA compiler version:
Cuda compilation tools, release 13.2, V13.2.86
Build cuda_13.2.r13.2/compiler.37953736_0