Add tests for bfloat16 math functions on CUDA #57984
Closed
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
🔗 Helpful links
❌ 5 New FailuresAs of commit 0b0a0c7 (more details on the Dr. CI page): Expand to see more
🕵️ 5 new failures recognized by patternsThe following CI failures do not appear to be due to upstream breakages
|
mruberry
added
the
triaged
ngimel
reviewed
May 12, 2021
aten/src/ATen/test/cuda_half_test.cu
Outdated
| @@ -51,7 +51,7 @@ __device__ void test(){ | |||
| assert(::abs(::acosh(Half(1.0)) - ::acosh(1.0f)) <= threshold); | |||
| assert(::abs(::acosh(Half(1.0)) - ::acosh(1.0f)) <= threshold); | |||
| assert(::abs(::asinh(Half(1.0)) - ::asinh(1.0f)) <= threshold); | |||
| assert(::abs(::atanh(Half(1.0)) - ::atanh(1.0f)) <= threshold); | |||
| assert(::abs(::atanh(Half(0.5)) - ::atanh(0.5f)) <= threshold); | |||
There was a problem hiding this comment.
I don't know. Very strange.
|
|
||
| using namespace at; | ||
|
|
||
| __device__ void test(){ |
There was a problem hiding this comment.
should this be a templated test rather than copy-paste? You can have BFloat16, __nb_bfloat16, __float2bfloat16 and their half counterparts as template parameters, is there anything else?
|
This seems fine. @ngimel do you want to make the final call? |
ngimel
approved these changes
May 12, 2021
|
@ngimel has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
CI Flow Status⚛️ CI FlowRuleset - Version:
You can add a comment to the PR and tag @pytorchbot with the following commands: # ciflow rerun, "ciflow/default" will always be added automatically
@pytorchbot ciflow rerun
# ciflow rerun with additional labels "-l <ciflow/label_name>", which is equivalent to adding these labels manually and trigger the rerun
@pytorchbot ciflow rerun -l ciflow/scheduled -l ciflow/slowFor more information, please take a look at the CI Flow Wiki. |
|
@ngimel This was not merged because a CUDA 11.0 bug causes internal build failure. Is the internal build still using CUDA 11.0 today? |
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
|
removing |
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
|
@pytorchbot rebase |
|
@pytorchbot successfully started a rebase job. Check the current status here |
|
Rebase failed due to Command Raised by https://github.com/pytorch/pytorch/actions/runs/2708457304 |
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/57984
Note: Links to docs will display an error until the docs builds have been completed. ❌ 6 Failures, 1 PendingAs of commit 2a79341: The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
/easycla As part of the transition to the PyTorch Foundation, this project now requires contributions be covered under the new CLA. See #85559 for additional details. This comment will trigger a new check of this PR. If you are already covered, you will simply see a new "EasyCLA" check that passes. If you are not covered, a bot will leave a new comment with a link to sign. |
|
@pytorchbot rebase |
|
@pytorchbot successfully started a rebase job. Check the current status here |
|
Rebase failed due to Command Raised by https://github.com/pytorch/pytorch/actions/runs/3527515860 |
|
@ngimel @zasdfgbnm Do we still want to merge this? |
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
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.
I copy paste
cuda_half_test.cutocuda_bfloat16_test.cuand change it to test bfloat16. It do find a few issues:CUDA_VERSIONsometimes doesn't work on c10 (I don't know the reason), I changed it to use__CUDACC_VER_MAJOR__insteadoperator __nv_bfloat16()ofc10::BFloat16should not be explicit-D__CUDA_NO_BFLOAT16_OPERATORS__to be consistent with half behaviorassert(::abs(::atanh(Half(1.0)) - ::atanh(1.0f)) <= threshold);in the test, this doesn't make sense, becauseatanh(1)is inf.