UCT/GGA: filter out unsupported resources from component list #10372
Conversation
This file contains 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
tvegas1
reviewed
Dec 13, 2024
iyastreb
reviewed
Dec 13, 2024
tvegas1
previously approved these changes
Dec 13, 2024
iyastreb
previously approved these changes
Dec 13, 2024
yosefe
reviewed
Dec 15, 2024
src/uct/ib/mlx5/gga/gga_mlx5.c
Outdated
| } | ||
|
|
||
| for (num_resources = 0, i = 0; i < num_ib_resources; ++i) { | ||
| status = uct_ib_mlx5_gga_md_open(component, resources[i].md_name, |
There was a problem hiding this comment.
opening the md during query is a too heavy operation, can we just do mlx5dv_open_device and call UCT_IB_MLX5_CMD_OP_QUERY_HCA_CAP? perhaps extract some common code with uct_ib_mlx5_devx_md_open to a helper function.
There was a problem hiding this comment.
AFAIU mlx5dv_open_device is a heaviest operation of md_open, do you think such refactoring makes sense to avoid a couple of malloc/free?
src/uct/ib/mlx5/gga/gga_mlx5.c
Outdated
Comment on lines
849
to
857
| mlx5_md = ucs_derived_of(md, uct_ib_mlx5_md_t); | ||
| if (!ucs_test_all_flags(mlx5_md->flags, UCT_GGA_MLX5_MD_CAPS)) { | ||
| ucs_debug("device %s does not match capabilities " | ||
| "required for GGA(%"PRIx32"), md_flags=%"PRIx32, | ||
| md_name, UCT_GGA_MLX5_MD_CAPS, mlx5_md->flags); | ||
| status = UCS_ERR_NO_DEVICE; | ||
| uct_ib_mlx5_devx_md_close(&md->super); | ||
| goto out_free_dev_list; | ||
| } |
evgeny-leksikov
dismissed stale reviews from iyastreb and tvegas1
via
cc9a420
yosefe
reviewed
Dec 19, 2024
Comment on lines
-727
to
-729
| !ucs_test_all_flags(mlx5_md->flags, UCT_IB_MLX5_MD_FLAG_DEVX | | ||
| UCT_IB_MLX5_MD_FLAG_INDIRECT_XGVMI | | ||
| UCT_IB_MLX5_MD_FLAG_MMO_DMA)) { |
|
/azp run |
|
Azure Pipelines successfully started running 4 pipeline(s). |
yosefe
approved these changes
Jan 1, 2025
|
/azp run perf |
|
Azure Pipelines successfully started running 1 pipeline(s). |
evgeny-leksikov
force-pushed
the
uct_gga_res_list
branch
from
f7ddc81 to
777757e
Compare
tvegas1
approved these changes
Jan 6, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
What?
Filter out unsupported resources from GGA component
Why?
full log: https://dev.azure.com/ucfconsort/0b36e3f0-8ab9-4a48-b68b-4b2350e02c88/_apis/build/builds/92757/logs/1012
How?
check capabilities