Use dcgm v4 on HWO #466
Merged
Use dcgm v4 on HWO #466
Conversation
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
- automatic dedetection to use the right channel for dcgm based on driver version - check if dcgm channel is valid
gabrielcocenza
requested review from
Deezzir,
H-M-Quang-Ngo,
Pjack,
chanchiwai-ray and
jneo8
This was referenced Sep 26, 2025
- check for comparibility of v3 and v4
Pjack
reviewed
Sep 30, 2025
jneo8
requested changes
Sep 30, 2025
- increase coverage
the driver is not installed or loaded
jneo8
previously approved these changes
Oct 1, 2025
- add comments - adapt unit tests
gabrielcocenza
force-pushed
the
dcgm_v4
branch
from
31419c1 to
e4e50fe
Compare
Pjack
requested changes
Oct 2, 2025
gabrielcocenza
force-pushed
the
dcgm_v4
branch
2 times, most recently
from
a7b250a to
8cf59a6
Compare
dcgm_v3_compatible - automatic channel selection just for auto
gabrielcocenza
force-pushed
the
dcgm_v4
branch
from
8cf59a6 to
3826007
Compare
chanchiwai-ray
approved these changes
Oct 3, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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Outdated suggestions cannot be applied.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.
Implementation of v4 and the possible dcgm snap channel configuration:
Automatic detection of dcgm when using
autoV3
V4
auto, however it would block a unit if cuda10 driver is detectedGeneral changes
checkin snap exporter. Discovered that sometimes when refreshing and checking in short period of time, services might not be ready yet which gives false status that the exporter is failingWith this change the default value will be set to
auto. With that existent deployments using v3 will be updated to the corresponding version on v4Note: The auto logic is installing from
edgebecause there isn't yetstablefor v4 tracksImplementation of : #465