You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2022/08/01 19:32:36 UTC
[GitHub] [tvm] masahi opened a new pull request, #12263: [MetaSchedule] Check auto tensorization applicability in MultiLevelTilingWithIntrin
masahi opened a new pull request, #12263:
URL: https://github.com/apache/tvm/pull/12263
Thanks for contributing to TVM! Please refer to guideline https://tvm.apache.org/docs/contribute/ for useful information and tips. After the pull request is submitted, please request code reviews from [Reviewers](https://github.com/apache/incubator-tvm/blob/master/CONTRIBUTORS.md#reviewers) by @ them in the pull request thread.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] vinx13 commented on a diff in pull request #12263: [MetaSchedule] Check auto tensorization applicability in MultiLevelTilingWithIntrin
Posted by GitBox <gi...@apache.org>.
vinx13 commented on code in PR #12263:
URL: https://github.com/apache/tvm/pull/12263#discussion_r934890744
##########
tests/python/unittest/test_meta_schedule_schedule_rule_multi_level_tiling.py:
##########
@@ -563,6 +563,53 @@ def test_multi_level_tiling_dense_dpa4():
check_trace(spaces, expected)
+def test_multi_level_tiling_dense_dp4a_non_tensorizable():
+ m, n, k = 128, 128, 128
Review Comment:
would be great to also add a test case of non-divisible shape (m=n=k=127, dtype=int8)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] vinx13 merged pull request #12263: [MetaSchedule] Check auto tensorization applicability in MultiLevelTilingWithIntrin
Posted by GitBox <gi...@apache.org>.
vinx13 merged PR #12263:
URL: https://github.com/apache/tvm/pull/12263
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] masahi commented on a diff in pull request #12263: [MetaSchedule] Check auto tensorization applicability in MultiLevelTilingWithIntrin
Posted by GitBox <gi...@apache.org>.
masahi commented on code in PR #12263:
URL: https://github.com/apache/tvm/pull/12263#discussion_r934994447
##########
tests/python/unittest/test_meta_schedule_schedule_rule_multi_level_tiling.py:
##########
@@ -560,7 +563,12 @@ def test_multi_level_tiling_dense_dpa4():
)
]
- check_trace(spaces, expected)
+ _test_multi_level_tiling_dense_dp4a(m, n, k, "int8", "int32", expected)
+
+
+def test_multi_level_tiling_dense_dp4a_non_tensorizable():
+ _test_multi_level_tiling_dense_dp4a(128, 128, 128, "float32", "float32", [""])
+ _test_multi_level_tiling_dense_dp4a(127, 127, 127, "int32", "int32", [""])
Review Comment:
Sorry this was a typo, I wanted to have `"int8", "int32"`. To my surprise, I got `Check failed: (tiled_loop_rv.defined()) is false` error.
Fixing.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] vinx13 commented on a diff in pull request #12263: [MetaSchedule] Check auto tensorization applicability in MultiLevelTilingWithIntrin
Posted by GitBox <gi...@apache.org>.
vinx13 commented on code in PR #12263:
URL: https://github.com/apache/tvm/pull/12263#discussion_r934990690
##########
tests/python/unittest/test_meta_schedule_schedule_rule_multi_level_tiling.py:
##########
@@ -560,7 +563,12 @@ def test_multi_level_tiling_dense_dpa4():
)
]
- check_trace(spaces, expected)
+ _test_multi_level_tiling_dense_dp4a(m, n, k, "int8", "int32", expected)
+
+
+def test_multi_level_tiling_dense_dp4a_non_tensorizable():
+ _test_multi_level_tiling_dense_dp4a(128, 128, 128, "float32", "float32", [""])
+ _test_multi_level_tiling_dense_dp4a(127, 127, 127, "int32", "int32", [""])
Review Comment:
what about `_test_multi_level_tiling_dense_dp4a(127, 127, 127, "int8", "int32", [""])`? what happen if the dtype matches with the intrin and the only mismatch is the shape?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [tvm] masahi commented on a diff in pull request #12263: [MetaSchedule] Check auto tensorization applicability in MultiLevelTilingWithIntrin
Posted by GitBox <gi...@apache.org>.
masahi commented on code in PR #12263:
URL: https://github.com/apache/tvm/pull/12263#discussion_r935015575
##########
tests/python/unittest/test_meta_schedule_schedule_rule_multi_level_tiling.py:
##########
@@ -560,7 +563,12 @@ def test_multi_level_tiling_dense_dpa4():
)
]
- check_trace(spaces, expected)
+ _test_multi_level_tiling_dense_dp4a(m, n, k, "int8", "int32", expected)
+
+
+def test_multi_level_tiling_dense_dp4a_non_tensorizable():
+ _test_multi_level_tiling_dense_dp4a(128, 128, 128, "float32", "float32", [""])
+ _test_multi_level_tiling_dense_dp4a(127, 127, 127, "int32", "int32", [""])
Review Comment:
Fixed
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org