CORE: fix score update when only score given #779
Merged
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
|
bot:retest |
vspetrov
approved these changes
May 18, 2023
Sergei-Lebedev
force-pushed
the
topic/fix_score_coll_update
branch
from
8faf086 to
771ac8d
Compare
shimmybalsam
approved these changes
May 21, 2023
src/coll_score/ucc_coll_score.h
Outdated
| ucc_coll_score_alloc_from_str. | ||
| Return values: | ||
| UCC_OK - input alg_id can be correctly mapped to the "init" fn | ||
| UCC_ERR_NOT_SUPPORTED - CL/TL does allow changing algorithms ids for |
There was a problem hiding this comment.
typo - shouod be "doesn't" instead of "does"
| int mt_n); | ||
|
|
||
| ucc_status_t | ||
| ucc_coll_score_update_from_str(const char *str, |
There was a problem hiding this comment.
Once merged, needs to be updated in TL/SHM as well.
| @@ -34,7 +34,8 @@ UCC_TEST_F(test_score_update, non_overlap) | |||
| UCC_COLL_TYPE_BARRIER); | |||
| init_score(update, RLIST({RANGE(10, 20, 100), RANGE(30, 35, 1)}), | |||
| UCC_COLL_TYPE_BARRIER); | |||
| EXPECT_EQ(UCC_OK, ucc_coll_score_update(score, update, 0, NULL, 0)); | |||
| EXPECT_EQ(UCC_OK, ucc_coll_score_update(score, update, 0, NULL, 0, | |||
| UCC_COLL_TYPE_ALL)); | |||
There was a problem hiding this comment.
Will UCC_COLL_TYPE_ALL work for TL/SHM as well? Not all colls are supported there - will it fall back onto other TL?
manjugv
approved these changes
May 31, 2023
Sergei-Lebedev
force-pushed
the
topic/fix_score_coll_update
branch
2 times, most recently
from
f4d0aae to
fe29e9c
Compare
Sergei-Lebedev
force-pushed
the
topic/fix_score_coll_update
branch
from
fe29e9c to
d06f9d6
Compare
janjust
pushed a commit
to janjust/ucc
that referenced
this pull request
Jan 31, 2024
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
Fixes ucc score table when user provides score for TL/CL and only total score is given
e.g. UCC_TL_SHARP_TUNE=inf results in scatter, allgather, alltoall collectives (not supported by sharp) in UCC score table
with fix
ucc_team.c:471 UCC INFO ===== COLL_SCORE_MAP (team_id 32768, size 2) ===== ucc_coll_score_map.c:201 UCC INFO Allgather: ucc_coll_score_map.c:201 UCC INFO Host: {0..4095}:TL_UCP:10 {4K..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Allgatherv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Allreduce: ucc_coll_score_map.c:201 UCC INFO Host: {0..4095}:TL_SHARP:10 {4K..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Alltoall: ucc_coll_score_map.c:201 UCC INFO Host: {0..257}:TL_UCP:10 {258..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Alltoallv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Barrier: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Bcast: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Fanin: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Fanout: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Gather: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Gatherv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Reduce: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Reduce_scatter: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Reduce_scatterv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_coll_score_map.c:201 UCC INFO Scatterv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_UCP:10 ucc_team.c:474 UCC INFO ================================================without fix
ucc_team.c:472 UCC INFO ===== COLL_SCORE_MAP (team_id 32771, size 2) ===== ucc_coll_score_map.c:201 UCC INFO Allgather: ucc_coll_score_map.c:201 UCC INFO Host: {0..4095}:TL_SHARP:10 {4K..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Allgatherv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Allreduce: ucc_coll_score_map.c:201 UCC INFO Host: {0..4095}:TL_SHARP:10 {4K..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Alltoall: ucc_coll_score_map.c:201 UCC INFO Host: {0..257}:TL_SHARP:10 {258..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Alltoallv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Barrier: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Bcast: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Fanin: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Fanout: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Gather: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Gatherv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Reduce: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Reduce_scatter: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Reduce_scatterv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Scatter: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Scatterv: ucc_coll_score_map.c:201 UCC INFO Host: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Cuda: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO CudaManaged: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO Rocm: {0..inf}:TL_SHARP:10 ucc_coll_score_map.c:201 UCC INFO RocmManaged: {0..inf}:TL_SHARP:10 ucc_team.c:474 UCC INFO ================================================How ?
Pass TL supported collectives to tuning string parser