[release/2.6] [SWDEV-529824] Fix Float16 CooperativeReduction Test Failure #2204
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
pmaybank
changed the title
pmaybank
changed the title
jeffdaily
approved these changes
May 29, 2025
Author
|
!cherry-pick --onto release/2.7 |
okakarpa
pushed a commit
that referenced
this pull request
May 30, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]>
Collaborator
|
Created branch autogenerated/release/2.7_cherry-pick_pr-2204 and #2211 |
Author
|
!cherry-pick --onto rocm6.5_internal_testing |
okakarpa
pushed a commit
that referenced
this pull request
May 30, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]>
Collaborator
|
Created branch autogenerated/rocm6.5_internal_testing_cherry-pick_pr-2204 and #2212 |
jithunnair-amd
pushed a commit
that referenced
this pull request
Jun 5, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]> (cherry picked from commit 8fe3cdd)
jithunnair-amd
pushed a commit
that referenced
this pull request
Jun 6, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]> (cherry picked from commit 8fe3cdd)
pruthvistony
pushed a commit
that referenced
this pull request
Jun 6, 2025
…6 CooperativeReduction Test Failure (#2211) Cherry-pick of #2204 Co-authored-by: pmaybank <[email protected]> Co-authored-by: pnikolic-amd <[email protected]>
pruthvistony
pushed a commit
that referenced
this pull request
Jun 6, 2025
…4] Fix Float16 CooperativeReduction Test Failure (#2212) Cherry-pick of #2204 Co-authored-by: pmaybank <[email protected]> Co-authored-by: pnikolic-amd <[email protected]>
Collaborator
|
!cherry-pick --onto rocm7.0_internal_testing |
okakarpa
pushed a commit
that referenced
this pull request
Jun 7, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]>
Collaborator
|
Created branch autogenerated/rocm7.0_internal_testing_cherry-pick_pr-2204 and #2252 |
jithunnair-amd
pushed a commit
that referenced
this pull request
Jun 7, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]>
pragupta
pushed a commit
that referenced
this pull request
Jul 21, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]> (cherry picked from commit 8fe3cdd) (cherry picked from commit 34f3b3e)
pragupta
pushed a commit
to pragupta/pytorch
that referenced
this pull request
Jul 21, 2025
…ilure (ROCm#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]> (cherry picked from commit 8fe3cdd) (cherry picked from commit 34f3b3e)
pragupta
pushed a commit
that referenced
this pull request
Jul 22, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]> (cherry picked from commit 8fe3cdd) (cherry picked from commit 34f3b3e)
jithunnair-amd
pushed a commit
that referenced
this pull request
Jul 22, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]> (cherry picked from commit 8fe3cdd) (cherry picked from commit 34f3b3e)
pragupta
pushed a commit
that referenced
this pull request
Jul 29, 2025
…ilure (#2204) - Previously expected values were calculated on GPU using same dtype as result values - Now expected values are calculated on CPU using Float32 dtype - This fixes a test failure that was observed on Navi48 where difference between Eager mode (expected) and Inductor / Triton (result) did not meet the error tolerance when sum was evaluated on an array of Float16 values Co-authored-by: pnikolic-amd <[email protected]> (cherry picked from commit 8fe3cdd) (cherry picked from commit 34f3b3e)
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.
Cherry-picked to release/2.7 branch via #2211
Cherry-picked to rocm6.5_internal_testing branch via #2212
Cherry-picked to rocm7.0_internal_testing branch via #2252