fix: NVML memory query fallback for DGX Spark#2203
Open
dbuos wants to merge 2 commits intoNVIDIA-NeMo:mainfrom
Open
fix: NVML memory query fallback for DGX Spark#2203dbuos wants to merge 2 commits intoNVIDIA-NeMo:mainfrom
dbuos wants to merge 2 commits intoNVIDIA-NeMo:mainfrom
Conversation
dbuos
force-pushed
the
fix/nvml-dgx-spark-fallback
branch
from
d3d46a9 to
fa4ec27
Compare
…supported nvmlDeviceGetMemoryInfo returns NVML_ERROR_NOT_SUPPORTED on DGX Spark (GB10). Log the error and fall back to torch.cuda.mem_get_info which works on all CUDA devices. Signed-off-by: Daniel Bustamante Ospina <dbustamante70@gmail.com>
dbuos
force-pushed
the
fix/nvml-dgx-spark-fallback
branch
from
db8ae12 to
07a4c07
Compare
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
Running GRPO training on NVIDIA DGX Spark (
examples/run_grpo.py) crashes during weight refit becausenvmlDeviceGetMemoryInfois not supported on GB10:Tested with the latest
nvidia-ml-py==13.595.45— same error.Fix
Fall back to
torch.cuda.mem_get_info()when the NVML call fails. This API works on all CUDA devices including DGX Spark.