Add CUDA 13.0 Tests for CuFile I/O Operations - #1060
Merged
Conversation
Contributor
Member
|
Thanks, Chloe! Pinning you internally... |
leofang
added
P0
leofang
requested changes
Oct 14, 2025
leofang
reviewed
Oct 14, 2025
Member
|
/ok to test e1eabf8 |
leofang
requested changes
Oct 17, 2025
Comment on lines
+1954
to
+1955
| # Reset to level 0 (disabled) for cleanup | ||
| cufile.set_stats_level(0) |
| @pytest.mark.skipif( | ||
| cufileVersionLessThan(1150), reason="cuFile parameter APIs require cuFile library version 13.0 or later" | ||
| ) | ||
| def test_stats_start(): |
Member
There was a problem hiding this comment.
@chloechia4 Please remove this test entirely. I don't see any change?
Comment on lines
+2025
to
+2026
| # Set statistics level first (required before starting stats) | ||
| cufile.set_stats_level(1) # Level 1 = basic statistics |
Member
There was a problem hiding this comment.
Where...? @chloechia4 if you made the changes locally, make sure you push it to remote.
Comment on lines
+2062
to
+2063
| # Set statistics level first (required before starting stats) | ||
| cufile.set_stats_level(1) # Level 1 = basic statistics |
Comment on lines
+2106
to
+2107
| # Reset cuFile statistics to clear all counters | ||
| cufile.stats_reset() |
| fd = os.open(file_path, os.O_CREAT | os.O_RDWR | os.O_DIRECT, 0o600) | ||
|
|
||
| try: | ||
| cufile.set_stats_level(2) # L2 = detailed performance metrics |
Contributor
Author
There was a problem hiding this comment.
Do you mean reset(). Adding a stats_reset() in "finally" section
Contributor
|
/ok to test |
Contributor
@mdboom, there was an error processing your request: See the following link for more information: https://docs.gha-runners.nvidia.com/cpr/e/1/ |
Contributor
|
/ok to test 65573fc |
Contributor
|
/ok to test 642be44 |
Contributor
|
/ok to test a9584c8 |
Contributor
@mdboom, there was an error processing your request: See the following link for more information: https://docs.gha-runners.nvidia.com/cpr/e/2/ |
Contributor
|
/ok to test a9584c8 |
Contributor
|
/ok to test 3f81345 |
Contributor
|
/ok to test ed6eeb2 |
mdboom
enabled auto-merge (squash)
October 20, 2025 18:26
leofang
approved these changes
Oct 20, 2025
mdboom
approved these changes
Oct 20, 2025
Contributor
|
/ok to test 125f968 |
|
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
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.
Description
This PR includes tests for the low-level bindings and the generated low-level bindings introduced in CUDA 13.0 for CUFile.
CUDA 13.0 CuFile Operations
Note: The original
test_batch_io_large_operations()did not pass once switched from CUDA 12.9 to 13.0. I realized it was because the file reads were occurring before the writes as it was submitting all operations (reads and writes) together in one batch. As a result, it was trivially failing because the reads would return as 0 bytes, since they were happening before any write I/O occurred. I changed it to so it would be separated into two phases: writes complete first in one batch handle, and then reads are submitted in another batch handle. This new test works with CUDA 12.9 versioning as well.All tests passing across CUDA versions
