Make it an error to TORCH_LIBRARY the same namespace twice. by ezyang · Pull Request #36505 · pytorch/pytorch

ezyang · 2020-04-13T20:22:55Z

Stack from ghstack:

Functionalize qconcat and register with new registration API #36536 Functionalize qconcat and register with new registration API
Convert qclamp to new operator registration API. #36535 Convert qclamp to new operator registration API.
Convert qbatch_norm to new operator registration API. #36534 Convert qbatch_norm to new operator registration API.
Update quantized README for registering operators with new API. #36531 Update quantized README for registering operators with new API.
Functionalize qadd and register with new registration API. #36527 Functionalize qadd and register with new registration API.
Add a centralized TORCH_LIBRARY declaration for quantized and xnnpack ops. #36520 Add a centralized TORCH_LIBRARY declaration for quantized and xnnpack ops.
Add explicit schema for quantized conv/conv_prepack #36513 Add explicit schema for quantized conv/conv_prepack
Port detach/detach to new registration API #36512 Port detach/detach to new registration API
Make it an error to TORCH_LIBRARY the same namespace twice. #36505 Make it an error to TORCH_LIBRARY the same namespace twice.
Port Resize to use new registration API. #36390 Port Resize to use new registration API.
Port to new registration API (part 1) #36389 Update reference to RegisterOperators in error message in Convolution
Propagate and report the file/line number of registrations on errors. #36264 Propagate and report the file/line number of registrations on errors.
Switch to pybind11 style registration function API. #36258 Switch to pybind11 style registration function API.
Unconditionally register schema even for manual registration. #36250 Unconditionally register schema even for manual registration.

This is implemented as an extra level of state and error checking in
Dispatcher, since Dispatcher is where put all the other state
(technically, nothing in Dispatcher needs this error check; it's
a higher level concern.)

Signed-off-by: Edward Z. Yang ezyang@fb.com

Differential Revision: D21002535

This is implemented as an extra level of state and error checking in Dispatcher, since Dispatcher is where put all the other state (technically, nothing in Dispatcher needs this error check; it's a higher level concern.) Signed-off-by: Edward Z. Yang <ezyang@fb.com> [ghstack-poisoned]

dr-ci · 2020-04-13T20:24:03Z

💊 Build failures summary and remediations

As of commit 5b514f5 (more details on the Dr. CI page):

3/4 failures introduced in this PR
1/4 broken upstream at merge base fb70b4f from Apr 13 until Apr 14 (4 commits; 8f501f3 - ed2d1cb)
Please rebase on the viable/strict branch (expand for instructions)

If your commit is newer than viable/strict, you can try basing on an older, stable commit:
```
git fetch https://github.com/pytorch/pytorch viable/strict
git rebase --onto FETCH_HEAD $(git merge-base origin/master HEAD)
```
If your commit is older than viable/strict:
```
git fetch https://github.com/pytorch/pytorch viable/strict
git rebase FETCH_HEAD
```
Check out the recency history of this "viable master" tracking branch.

🕵️ 3 new failures recognized by patterns

The following build failures do not appear to be due to upstream breakages:

pytorch_linux_xenial_py3_clang5_mobile_custom_build_static (1/3)

Step: "Build" (full log | pattern match details | 🔁 rerun)

Apr 14 13:47:04 caused by: Connection refused (os error 111)

Apr 14 13:47:04 +++ eval 'extract_trap_cmd ' 
Apr 14 13:47:04 ++++ extract_trap_cmd 
Apr 14 13:47:04 ++++ printf '%s\n' '' 
Apr 14 13:47:04 +++ printf '%s\n' cleanup 
Apr 14 13:47:04 ++ trap -- ' 
Apr 14 13:47:04 cleanup' EXIT 
Apr 14 13:47:04 ++ which sccache 
Apr 14 13:47:04 ++ sccache --stop-server 
Apr 14 13:47:04 Stopping sccache server... 
Apr 14 13:47:04 error: couldn't connect to server 
Apr 14 13:47:04 caused by: Connection refused (os error 111) 
Apr 14 13:47:04 ++ true 
Apr 14 13:47:04 ++ rm /var/lib/jenkins/sccache_error.log 
Apr 14 13:47:04 rm: cannot remove '/var/lib/jenkins/sccache_error.log': No such file or directory 
Apr 14 13:47:04 ++ true 
Apr 14 13:47:04 ++ SCCACHE_ERROR_LOG=/var/lib/jenkins/sccache_error.log 
Apr 14 13:47:04 ++ SCCACHE_IDLE_TIMEOUT=1200 
Apr 14 13:47:04 ++ RUST_LOG=sccache::server=error 
Apr 14 13:47:04 ++ sccache --start-server 
Apr 14 13:47:04 Starting sccache server... 
Apr 14 13:47:04 ++ sccache --zero-stats

pytorch_linux_xenial_py3_clang5_mobile_custom_build_dynamic (2/3)

Step: "Build" (full log | pattern match details | 🔁 rerun)

Apr 14 13:49:05 caused by: Connection refused (os error 111)

Apr 14 13:49:05 +++ eval 'extract_trap_cmd ' 
Apr 14 13:49:05 ++++ extract_trap_cmd 
Apr 14 13:49:05 ++++ printf '%s\n' '' 
Apr 14 13:49:05 +++ printf '%s\n' cleanup 
Apr 14 13:49:05 ++ trap -- ' 
Apr 14 13:49:05 cleanup' EXIT 
Apr 14 13:49:05 ++ which sccache 
Apr 14 13:49:05 ++ sccache --stop-server 
Apr 14 13:49:05 Stopping sccache server... 
Apr 14 13:49:05 error: couldn't connect to server 
Apr 14 13:49:05 caused by: Connection refused (os error 111) 
Apr 14 13:49:05 ++ true 
Apr 14 13:49:05 ++ rm /var/lib/jenkins/sccache_error.log 
Apr 14 13:49:05 rm: cannot remove '/var/lib/jenkins/sccache_error.log': No such file or directory 
Apr 14 13:49:05 ++ true 
Apr 14 13:49:05 ++ SCCACHE_ERROR_LOG=/var/lib/jenkins/sccache_error.log 
Apr 14 13:49:05 ++ SCCACHE_IDLE_TIMEOUT=1200 
Apr 14 13:49:05 ++ RUST_LOG=sccache::server=error 
Apr 14 13:49:05 ++ sccache --start-server 
Apr 14 13:49:05 Starting sccache server... 
Apr 14 13:49:05 ++ sccache --zero-stats

pytorch_linux_xenial_py3_6_gcc5_4_build (3/3)

Step: "Build" (full log | pattern match details | 🔁 rerun)

Apr 14 14:57:03 /var/lib/jenkins/workspace/aten/src/ATen/core/op_registration/op_registration_test.cpp:1545:22: error: 'import' is not a member of 'c10'

Apr 14 14:57:01 [ 92%] Linking CXX executable ../bin/SubgraphMatcherTest 
Apr 14 14:57:01 [ 92%] Linking CXX executable ../bin/AlgorithmsTest 
Apr 14 14:57:01 [ 92%] Built target SubgraphMatcherTest 
Apr 14 14:57:01 Scanning dependencies of target elementwise_op_test 
Apr 14 14:57:01 [ 92%] Building CXX object caffe2/CMakeFiles/elementwise_op_test.dir/operators/elementwise_op_test.cc.o 
Apr 14 14:57:02 [ 92%] Built target AlgorithmsTest 
Apr 14 14:57:02 Scanning dependencies of target GraphTest 
Apr 14 14:57:02 [ 92%] Building CXX object caffe2/CMakeFiles/GraphTest.dir/core/nomnigraph/tests/GraphTest.cc.o 
Apr 14 14:57:03 [ 92%] Linking CXX executable ../bin/GraphTest 
Apr 14 14:57:03 /var/lib/jenkins/workspace/aten/src/ATen/core/op_registration/op_registration_test.cpp: In member function 'virtual void {anonymous}::OperatorRegistrationTest_whenDeregisteringOp_thenHandleBecomesInvalid_Test::TestBody()': 
Apr 14 14:57:03 /var/lib/jenkins/workspace/aten/src/ATen/core/op_registration/op_registration_test.cpp:1545:22: error: 'import' is not a member of 'c10' 
Apr 14 14:57:03      auto registrar = c10::import() 
Apr 14 14:57:03                       ^ 
Apr 14 14:57:03 make[2]: *** [caffe2/CMakeFiles/op_registration_test.dir/__/aten/src/ATen/core/op_registration/op_registration_test.cpp.o] Error 1 
Apr 14 14:57:03 caffe2/CMakeFiles/op_registration_test.dir/build.make:62: recipe for target 'caffe2/CMakeFiles/op_registration_test.dir/__/aten/src/ATen/core/op_registration/op_registration_test.cpp.o' failed 
Apr 14 14:57:03 make[1]: *** [caffe2/CMakeFiles/op_registration_test.dir/all] Error 2 
Apr 14 14:57:03 make[1]: *** Waiting for unfinished jobs.... 
Apr 14 14:57:03 CMakeFiles/Makefile2:8480: recipe for target 'caffe2/CMakeFiles/op_registration_test.dir/all' failed 
Apr 14 14:57:03 [ 92%] Built target GraphTest 
Apr 14 14:57:09 [ 92%] Linking CXX executable ../bin/elementwise_op_test 
Apr 14 14:57:10 [ 92%] Built target elementwise_op_test

🚧 1 upstream failure:

These were probably caused by upstream breakages:

pytorch_xla_linux_xenial_py3_6_clang7_build from Apr 13 until Apr 14 (4 commits; 8f501f3 - ed2d1cb)
- 🔁 rerun

This comment was automatically generated by Dr. CI (expand for details).

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions on the GitHub issue tracker.

See how this bot performed.

This comment has been revised 23 times.

This is implemented as an extra level of state and error checking in Dispatcher, since Dispatcher is where put all the other state (technically, nothing in Dispatcher needs this error check; it's a higher level concern.) Signed-off-by: Edward Z. Yang <ezyang@fb.com> [ghstack-poisoned]

This is implemented as an extra level of state and error checking in Dispatcher, since Dispatcher is where put all the other state (technically, nothing in Dispatcher needs this error check; it's a higher level concern.) Signed-off-by: Edward Z. Yang <ezyang@fb.com> Differential Revision: [D21002535](https://our.internmc.facebook.com/intern/diff/D21002535) [ghstack-poisoned]

This is implemented as an extra level of state and error checking in Dispatcher, since Dispatcher is where put all the other state (technically, nothing in Dispatcher needs this error check; it's a higher level concern.) Signed-off-by: Edward Z. Yang <ezyang@fb.com> ghstack-source-id: 3bb3e5a Pull Request resolved: pytorch#36505

ezyang · 2020-04-15T17:45:52Z

Folded into #36258

This is implemented as an extra level of state and error checking in Dispatcher, since Dispatcher is where put all the other state (technically, nothing in Dispatcher needs this error check; it's a higher level concern.) Signed-off-by: Edward Z. Yang <ezyang@fb.com> ghstack-source-id: 3bb3e5a Pull Request resolved: pytorch/pytorch#36505

This was referenced Apr 13, 2020

Port detach/detach to new registration API #36512

Closed

Add explicit schema for quantized conv/conv_prepack #36513

Closed

ezyang added 3 commits April 13, 2020 14:07

This was referenced Apr 13, 2020

Add a centralized TORCH_LIBRARY declaration for quantized and xnnpack ops. #36520

Closed

Functionalize qadd and register with new registration API. #36527

Closed

ezyang requested review from bhosmer and smessmer and removed request for smessmer April 13, 2020 22:42

ezyang closed this Apr 15, 2020

facebook-github-bot deleted the gh/ezyang/718/head branch May 16, 2020 14:16

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Make it an error to TORCH_LIBRARY the same namespace twice.#36505

Make it an error to TORCH_LIBRARY the same namespace twice.#36505
ezyang wants to merge 6 commits intogh/ezyang/718/basefrom
gh/ezyang/718/head

ezyang commented Apr 13, 2020 •

edited

Loading

Uh oh!

dr-ci Bot commented Apr 13, 2020 •

edited

Loading

Uh oh!

ezyang commented Apr 15, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

ezyang commented Apr 13, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dr-ci Bot commented Apr 13, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

💊 Build failures summary and remediations

🕵️ 3 new failures recognized by patterns

pytorch_linux_xenial_py3_clang5_mobile_custom_build_static (1/3)

pytorch_linux_xenial_py3_clang5_mobile_custom_build_dynamic (2/3)

pytorch_linux_xenial_py3_6_gcc5_4_build (3/3)

🚧 1 upstream failure:

Uh oh!

ezyang commented Apr 15, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

ezyang commented Apr 13, 2020 •

edited

Loading

dr-ci Bot commented Apr 13, 2020 •

edited

Loading