Uh oh!
There was an error while loading. Please reload this page.
[UnitTests] Added cuDNN to default test targets - #8383
Merged
Conversation
Lunderbergforce-pushed
the
tvm_testing_default_targets
branch
from
July 8, 2021 16:13
04a36d4 to
1a2838eCompareLunderberg
commented
Jul 8, 2021
ContributorAuthor
Updated to fix pylint error. |
Lunderbergforce-pushed
the
tvm_testing_default_targets
branch
2 times, most recently
from
July 12, 2021 16:14
b6fb4d9 to
f57debbCompareLunderberg
commented
Jul 12, 2021
ContributorAuthor
Current failures.
|
Lunderbergforce-pushed
the
tvm_testing_default_targets
branch
from
July 22, 2021 13:47
f57debb to
0ab1190CompareLunderberg
requested review from
ZihengJiang, areusch, comaniac, jroesch, junrushao, kparzysz-quic, masahi, merrymercy, tqchen, vinx13, yzhliu and zhiics
as code ownersJuly 22, 2021 13:47
comaniac
approved these changes
Jul 22, 2021
comaniac
left a comment
Contributor
There was a problem hiding this comment.
It would be better to also let a Vulkan folk review the changes in target_kind.cc.
Comment on lines
+383
to
+384
| if len(target_names) == 0: | ||
| target_names = DEFAULT_TEST_TARGETS |
Contributor
There was a problem hiding this comment.
nit
Suggested change
| iflen(target_names) ==0: | |
| target_names=DEFAULT_TEST_TARGETS | |
| ifnottarget_names: | |
| target_names=DEFAULT_TEST_TARGETS |
ContributorAuthor
There was a problem hiding this comment.
Good call, and updated.
Lunderbergforce-pushed
the
tvm_testing_default_targets
branch
4 times, most recently
from
July 29, 2021 15:16
80db1a3 to
4df4417CompareLunderbergforce-pushed
the
tvm_testing_default_targets
branch
from
August 3, 2021 14:43
4df4417 to
93a9535CompareLunderberg
commented
Aug 3, 2021
ContributorAuthor
…arget - Read target.kind.name instead of using string manipulation. - Target device query on a non-existent target is no longer an error. This occurs if expanding `vulkan -from_device=0` on a non-GPU machine.
Some unit tests explicitly test cudnn in addition to `tvm.testing.enabled_targets()`. This moved the cudnn checks into the same framework as all other targets, and adds it to the default list of targets to be run. Also, added `@tvm.testing.requires_cudnn` for tests specific to cudnn.
Lunderbergforce-pushed
the
tvm_testing_default_targets
branch
from
August 6, 2021 20:41
93a9535 to
3d2d719CompareLunderberg
requested review from
Huyuwei, Laurawly, jcf94, jwfromm, kevinthesun and mbrookhart
as code ownersAugust 6, 2021 20:41
Lunderberg
commented
Aug 6, 2021
ContributorAuthor
Hopefully the last update needed here, rebased to main and fixed an issue with the conv2d_cudnn handling of dilation. |
Previously, cuda/nvptx targets were excluded. Changed it to look up by target.kind.name, and to also exclude vulkan/opencl, as the dynamic lookup currently doesn't work on those backends.
This was referenced Aug 9, 2021
mehrdadh pushed a commit
to mehrdadh/tvm
that referenced
this pull request
Aug 11, 2021
* [Target][UnitTests] Look up target requirements based on tvm.target.Target - Read target.kind.name instead of using string manipulation. - Target device query on a non-existent target is no longer an error. This occurs if expanding `vulkan -from_device=0` on a non-GPU machine. * [UnitTests] Added cuDNN target to default test targets Some unit tests explicitly test cudnn in addition to `tvm.testing.enabled_targets()`. This moved the cudnn checks into the same framework as all other targets, and adds it to the default list of targets to be run. Also, added `@tvm.testing.requires_cudnn` for tests specific to cudnn. * [UnitTests] pytest.xfail for CuDNN conv2d with asymmetric padding * [Topi][CuDNN] Added handling of dilation to conv2d_cudnn * [Topi] Skip dynamic batch matmul on cudnn, vulkan, opencl Previously, cuda/nvptx targets were excluded. Changed it to look up by target.kind.name, and to also exclude vulkan/opencl, as the dynamic lookup currently doesn't work on those backends. Co-authored-by: Eric Lunderberg <elunderberg@octoml.ai>
ylc pushed a commit
to ylc/tvm
that referenced
this pull request
Sep 29, 2021
* [Target][UnitTests] Look up target requirements based on tvm.target.Target - Read target.kind.name instead of using string manipulation. - Target device query on a non-existent target is no longer an error. This occurs if expanding `vulkan -from_device=0` on a non-GPU machine. * [UnitTests] Added cuDNN target to default test targets Some unit tests explicitly test cudnn in addition to `tvm.testing.enabled_targets()`. This moved the cudnn checks into the same framework as all other targets, and adds it to the default list of targets to be run. Also, added `@tvm.testing.requires_cudnn` for tests specific to cudnn. * [UnitTests] pytest.xfail for CuDNN conv2d with asymmetric padding * [Topi][CuDNN] Added handling of dilation to conv2d_cudnn * [Topi] Skip dynamic batch matmul on cudnn, vulkan, opencl Previously, cuda/nvptx targets were excluded. Changed it to look up by target.kind.name, and to also exclude vulkan/opencl, as the dynamic lookup currently doesn't work on those backends. Co-authored-by: Eric Lunderberg <elunderberg@octoml.ai>
ylc pushed a commit
to ylc/tvm
that referenced
this pull request
Jan 13, 2022
* [Target][UnitTests] Look up target requirements based on tvm.target.Target - Read target.kind.name instead of using string manipulation. - Target device query on a non-existent target is no longer an error. This occurs if expanding `vulkan -from_device=0` on a non-GPU machine. * [UnitTests] Added cuDNN target to default test targets Some unit tests explicitly test cudnn in addition to `tvm.testing.enabled_targets()`. This moved the cudnn checks into the same framework as all other targets, and adds it to the default list of targets to be run. Also, added `@tvm.testing.requires_cudnn` for tests specific to cudnn. * [UnitTests] pytest.xfail for CuDNN conv2d with asymmetric padding * [Topi][CuDNN] Added handling of dilation to conv2d_cudnn * [Topi] Skip dynamic batch matmul on cudnn, vulkan, opencl Previously, cuda/nvptx targets were excluded. Changed it to look up by target.kind.name, and to also exclude vulkan/opencl, as the dynamic lookup currently doesn't work on those backends. Co-authored-by: Eric Lunderberg <elunderberg@octoml.ai>
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 freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
Some unit tests explicitly test cudnn in addition to tvm.testing.enabled_targets(). This moved the cudnn checks into the same framework as all other targets, and adds it to the default list of targets to be run. Also, added
@tvm.testing.requires_cudnnfor tests specific to cudnn.In principle, this should pass CI, since any non-cudnn tests can fall back to the cuda implementation, but "should" is a bit of a magic word.