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 2021/10/01 16:41:07 UTC

[GitHub] [tvm] tkonolige opened a new pull request #9175: Fix end to end benchmark with rpc devices

tkonolige opened a new pull request #9175:
URL: https://github.com/apache/tvm/pull/9175


   I forgot to convert remote devices to local devices in the `benchmark` rpc call. This PR fixes the issue and also adds a test to catch it. I've also improved the error message for the failure.


-- 
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] areusch commented on a change in pull request #9175: Fix end to end benchmark with rpc devices

Posted by GitBox <gi...@apache.org>.
areusch commented on a change in pull request #9175:
URL: https://github.com/apache/tvm/pull/9175#discussion_r722541594



##########
File path: tests/python/relay/test_backend_graph_executor.py
##########
@@ -361,21 +361,21 @@ def test_benchmark_end_to_end(dev, target):
     assert len(result.results) == 2
 
 
-@tvm.testing.requires_llvm
+@tvm.testing.requires_cuda
 def test_benchmark_end_to_end_rpc():
     server = rpc.Server("127.0.0.1")
     remote = rpc.connect(server.host, server.port)
 
     mod, params = mlp.get_workload(1)
-    lib = relay.build(mod, target="llvm", params=params)
+    lib = relay.build(mod, target="cuda", params=params)

Review comment:
       ah gotcha.




-- 
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] areusch commented on a change in pull request #9175: Fix end to end benchmark with rpc devices

Posted by GitBox <gi...@apache.org>.
areusch commented on a change in pull request #9175:
URL: https://github.com/apache/tvm/pull/9175#discussion_r721803719



##########
File path: tests/python/relay/test_backend_graph_executor.py
##########
@@ -361,21 +361,21 @@ def test_benchmark_end_to_end(dev, target):
     assert len(result.results) == 2
 
 
-@tvm.testing.requires_llvm
+@tvm.testing.requires_cuda
 def test_benchmark_end_to_end_rpc():
     server = rpc.Server("127.0.0.1")
     remote = rpc.connect(server.host, server.port)
 
     mod, params = mlp.get_workload(1)
-    lib = relay.build(mod, target="llvm", params=params)
+    lib = relay.build(mod, target="cuda", params=params)

Review comment:
       why this change?




-- 
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] areusch merged pull request #9175: Fix end to end benchmark with rpc devices

Posted by GitBox <gi...@apache.org>.
areusch merged pull request #9175:
URL: https://github.com/apache/tvm/pull/9175


   


-- 
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] tkonolige commented on pull request #9175: Fix end to end benchmark with rpc devices

Posted by GitBox <gi...@apache.org>.
tkonolige commented on pull request #9175:
URL: https://github.com/apache/tvm/pull/9175#issuecomment-933894413


   @areusch or @junrushao1994 Can you review?


-- 
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] tkonolige commented on a change in pull request #9175: Fix end to end benchmark with rpc devices

Posted by GitBox <gi...@apache.org>.
tkonolige commented on a change in pull request #9175:
URL: https://github.com/apache/tvm/pull/9175#discussion_r722408684



##########
File path: tests/python/relay/test_backend_graph_executor.py
##########
@@ -361,21 +361,21 @@ def test_benchmark_end_to_end(dev, target):
     assert len(result.results) == 2
 
 
-@tvm.testing.requires_llvm
+@tvm.testing.requires_cuda
 def test_benchmark_end_to_end_rpc():
     server = rpc.Server("127.0.0.1")
     remote = rpc.connect(server.host, server.port)
 
     mod, params = mlp.get_workload(1)
-    lib = relay.build(mod, target="llvm", params=params)
+    lib = relay.build(mod, target="cuda", params=params)

Review comment:
       The check that prevents you from copying from local to remote only allows copying from local cpu to remote cpu, so this test was not having any issues even though its behavior was wrong. (https://github.com/apache/tvm/pull/9175/files#diff-c2d3e12882a0e96941e57203f5f4c40eb7d9cf2e913d991fe8307f0930c2bb82R234-R238).




-- 
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