[Pytoch][Vulkan] Create context for layernorm #114701

copyrightly · 2023-11-28T20:39:38Z

Summary:
Layernorm has two arguments weight and bias which are stored as constant tensors on the CPU and they are transferred to GPU at every inference call. We create a context for this op to avoid the repeated passing. Specifically, we

created create_layernorm_context and run_layernorm_context in Layernorm.h and Layernorm.cpp
registered them in Register.cpp
rewrote the graph representation of the op in vulkan_rewrite.cpp

Test Plan:

Numerical test

[luwei@devbig984.prn1 /data/users/luwei/fbsource (b6ccc956c)]$ LD_LIBRARY_PATH=third-party/swiftshader/lib/linux-x64/ buck run fbcode/mode/dev-nosan //xplat/caffe2:pt_vulkan_api_test_bin -- --gtest_filter="*layer_norm*"
Recommended: For faster builds try buck2: replace 'buck' with 'buck2'
NOTE: buck-out/ has changed: look for files in fbsource/buck-out/v2/
'buck2 build --show-output //xplat/caffe2:pt_vulkan_api_test_bin' will print the new output paths.


If you are building in fbsource//xplat and have questions, post in 'Cross Platform Dev Discussions': https://fb.workplace.com/groups/xplat.qa

  Targets matching .buckconfig buck2.supported_projects:
  {'//xplat/caffe2:pt_vulkan_api_test_bin': '//xplat'}

  To suppress this warning: touch ~/.config/.dont_hint_buck2

Building: finished in 0.1 sec (100%) 339/339 jobs, 0/339 updated
  Total time: 0.2 sec
BUILD SUCCEEDED
Running main() from third-party/googletest/1.14.0/googletest/googletest/src/gtest_main.cc
Note: Google Test filter = *layer_norm*
[==========] Running 10 tests from 1 test suite.
[----------] Global test environment set-up.
[----------] 10 tests from VulkanAPITest
[ RUN      ] VulkanAPITest.packed_layer_norm_2d
[       OK ] VulkanAPITest.packed_layer_norm_2d (342 ms)
[ RUN      ] VulkanAPITest.packed_layer_norm_3d
[       OK ] VulkanAPITest.packed_layer_norm_3d (284 ms)
[ RUN      ] VulkanAPITest.packed_layer_norm_4d
[       OK ] VulkanAPITest.packed_layer_norm_4d (5 ms)
[ RUN      ] VulkanAPITest.layer_norm_invalid_inputs
[       OK ] VulkanAPITest.layer_norm_invalid_inputs (28 ms)
[ RUN      ] VulkanAPITest.layer_norm_2d
[       OK ] VulkanAPITest.layer_norm_2d (1 ms)
[ RUN      ] VulkanAPITest.layer_norm_3d
[       OK ] VulkanAPITest.layer_norm_3d (2 ms)
[ RUN      ] VulkanAPITest.layer_norm_4d
[       OK ] VulkanAPITest.layer_norm_4d (4 ms)
[ RUN      ] VulkanAPITest.native_layer_norm_2d
[       OK ] VulkanAPITest.native_layer_norm_2d (1 ms)
[ RUN      ] VulkanAPITest.native_layer_norm_3d
[       OK ] VulkanAPITest.native_layer_norm_3d (2 ms)
[ RUN      ] VulkanAPITest.native_layer_norm_4d
[       OK ] VulkanAPITest.native_layer_norm_4d (6 ms)
[----------] 10 tests from VulkanAPITest (679 ms total)

[----------] Global test environment tear-down
[==========] 10 tests from 1 test suite ran. (679 ms total)
[  PASSED  ] 10 tests.

Full test result in P888496077, summary as below

[----------] 419 tests from VulkanAPITest (21652 ms total)

[----------] Global test environment tear-down
[==========] 419 tests from 1 test suite ran. (21652 ms total)
[  PASSED  ] 418 tests.
[  SKIPPED ] 1 test, listed below:
[  SKIPPED ] VulkanAPITest.querypool_flushed_shader_log

Graph representation comparison

We created a model using layer_norm and traced it as below

class MyModel(torch.nn.Module):
    def __init__(self):
        super(MyModel, self).__init__()
        self.layer_norm = torch.nn.LayerNorm(normalized_shape=10)

    def forward(self, x):
        return self.layer_norm(x)

# Create an instance of the model
model = MyModel()

# Create a dummy input tensor for tracing
input_tensor = torch.randn(1, 10)

# Use torch.jit.trace to trace the model and generate a graph
traced_model = torch.jit.trace(model, input_tensor)

Then we converted the traced model to Vulkan backend using optimize_for_mobile

from torch.utils import mobile_optimizer

vulkan_model = mobile_optimizer.optimize_for_mobile(
    traced_model, backend="vulkan", preserved_methods=to_preserve
)

Then we can print the graph of the vulkan_model as print(vk_model.graph)

Before this diff

  %4 : bool = prim::Constant[value=1](), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0
  %5 : float = prim::Constant[value=1.0000000000000001e-05](), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0
  %14 : int[] = prim::Constant[value=[10]]()
  %33 : Tensor = aten::to(%x, %53, %30, %31, %31)
  %10 : Tensor = aten::layer_norm(%33, %14, %self.layer_norm.weight, %self.layer_norm.bias, %5, %4), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0

after this diff

  %14 : int[] = prim::Constant[value=[10]]()
  %47 : Tensor = aten::to(%x, %78, %44, %45, %45)
  %16 : Tensor = vulkan_prepack::run_layernorm_context(%47, %14, %17)

Reviewed By: SS-JIA

Differential Revision: D51530478

pytorch-bot · 2023-11-28T20:39:42Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/114701

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit f96df3b with merge base e891a3b ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

facebook-github-bot · 2023-11-28T20:39:49Z

This pull request was exported from Phabricator. Differential Revision: D51530478

Summary: `Layernorm` has two arguments weight and bias which are stored as constant tensors on the CPU and they are transferred to GPU at every inference call. We create a context for this op to avoid the repeated passing. Specifically, we - created `create_layernorm_context` and `run_layernorm_context` in `Layernorm.h` and `Layernorm.cpp` - registered them in `Register.cpp` - rewrote the graph representation of the op in `vulkan_rewrite.cpp` Test Plan: ## Numerical test ``` [luwei@devbig984.prn1 /data/users/luwei/fbsource (b6ccc956c)]$ LD_LIBRARY_PATH=third-party/swiftshader/lib/linux-x64/ buck run fbcode/mode/dev-nosan //xplat/caffe2:pt_vulkan_api_test_bin -- --gtest_filter="*layer_norm*" Recommended: For faster builds try buck2: replace 'buck' with 'buck2' NOTE: buck-out/ has changed: look for files in fbsource/buck-out/v2/ 'buck2 build --show-output //xplat/caffe2:pt_vulkan_api_test_bin' will print the new output paths. If you are building in fbsource//xplat and have questions, post in 'Cross Platform Dev Discussions': https://fb.workplace.com/groups/xplat.qa Targets matching .buckconfig buck2.supported_projects: {'//xplat/caffe2:pt_vulkan_api_test_bin': '//xplat'} To suppress this warning: touch ~/.config/.dont_hint_buck2 Building: finished in 0.1 sec (100%) 339/339 jobs, 0/339 updated Total time: 0.2 sec BUILD SUCCEEDED Running main() from third-party/googletest/1.14.0/googletest/googletest/src/gtest_main.cc Note: Google Test filter = *layer_norm* [==========] Running 10 tests from 1 test suite. [----------] Global test environment set-up. [----------] 10 tests from VulkanAPITest [ RUN ] VulkanAPITest.packed_layer_norm_2d [ OK ] VulkanAPITest.packed_layer_norm_2d (342 ms) [ RUN ] VulkanAPITest.packed_layer_norm_3d [ OK ] VulkanAPITest.packed_layer_norm_3d (284 ms) [ RUN ] VulkanAPITest.packed_layer_norm_4d [ OK ] VulkanAPITest.packed_layer_norm_4d (5 ms) [ RUN ] VulkanAPITest.layer_norm_invalid_inputs [ OK ] VulkanAPITest.layer_norm_invalid_inputs (28 ms) [ RUN ] VulkanAPITest.layer_norm_2d [ OK ] VulkanAPITest.layer_norm_2d (1 ms) [ RUN ] VulkanAPITest.layer_norm_3d [ OK ] VulkanAPITest.layer_norm_3d (2 ms) [ RUN ] VulkanAPITest.layer_norm_4d [ OK ] VulkanAPITest.layer_norm_4d (4 ms) [ RUN ] VulkanAPITest.native_layer_norm_2d [ OK ] VulkanAPITest.native_layer_norm_2d (1 ms) [ RUN ] VulkanAPITest.native_layer_norm_3d [ OK ] VulkanAPITest.native_layer_norm_3d (2 ms) [ RUN ] VulkanAPITest.native_layer_norm_4d [ OK ] VulkanAPITest.native_layer_norm_4d (6 ms) [----------] 10 tests from VulkanAPITest (679 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test suite ran. (679 ms total) [ PASSED ] 10 tests. ``` Full test result in P888496077, summary as below ``` [----------] 419 tests from VulkanAPITest (21652 ms total) [----------] Global test environment tear-down [==========] 419 tests from 1 test suite ran. (21652 ms total) [ PASSED ] 418 tests. [ SKIPPED ] 1 test, listed below: [ SKIPPED ] VulkanAPITest.querypool_flushed_shader_log ``` ## Graph representation comparison We created a model using `layer_norm` and traced it as below ``` class MyModel(torch.nn.Module): def __init__(self): super(MyModel, self).__init__() self.layer_norm = torch.nn.LayerNorm(normalized_shape=10) def forward(self, x): return self.layer_norm(x) # Create an instance of the model model = MyModel() # Create a dummy input tensor for tracing input_tensor = torch.randn(1, 10) # Use torch.jit.trace to trace the model and generate a graph traced_model = torch.jit.trace(model, input_tensor) ``` Then we converted the traced model to Vulkan backend using `optimize_for_mobile` ``` from torch.utils import mobile_optimizer vulkan_model = mobile_optimizer.optimize_for_mobile( traced_model, backend="vulkan", preserved_methods=to_preserve ) ``` Then we can print the graph of the `vulkan_model` as `print(vk_model.graph)` - Before this diff ``` %4 : bool = prim::Constant[value=1](), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0 %5 : float = prim::Constant[value=1.0000000000000001e-05](), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0 %14 : int[] = prim::Constant[value=[10]]() %33 : Tensor = aten::to(%x, %53, %30, %31, %31) %10 : Tensor = aten::layer_norm(%33, %14, %self.layer_norm.weight, %self.layer_norm.bias, %5, %4), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0 ``` - after this diff ``` %14 : int[] = prim::Constant[value=[10]]() %47 : Tensor = aten::to(%x, %78, %44, %45, %45) %16 : Tensor = vulkan_prepack::run_layernorm_context(%47, %14, %17) ``` Reviewed By: SS-JIA Differential Revision: D51530478

facebook-github-bot · 2023-11-29T04:24:00Z

This pull request was exported from Phabricator. Differential Revision: D51530478

Summary: `Layernorm` has two arguments weight and bias which are stored as constant tensors on the CPU and they are transferred to GPU at every inference call. We create a context for this op to avoid the repeated passing. Specifically, we - created `create_layernorm_context` and `run_layernorm_context` in `Layernorm.h` and `Layernorm.cpp` - registered them in `Register.cpp` - rewrote the graph representation of the op in `vulkan_rewrite.cpp` Test Plan: ## Numerical test ``` [luwei@devbig984.prn1 /data/users/luwei/fbsource (b6ccc956c)]$ LD_LIBRARY_PATH=third-party/swiftshader/lib/linux-x64/ buck run fbcode/mode/dev-nosan //xplat/caffe2:pt_vulkan_api_test_bin -- --gtest_filter="*layer_norm*" Recommended: For faster builds try buck2: replace 'buck' with 'buck2' NOTE: buck-out/ has changed: look for files in fbsource/buck-out/v2/ 'buck2 build --show-output //xplat/caffe2:pt_vulkan_api_test_bin' will print the new output paths. If you are building in fbsource//xplat and have questions, post in 'Cross Platform Dev Discussions': https://fb.workplace.com/groups/xplat.qa Targets matching .buckconfig buck2.supported_projects: {'//xplat/caffe2:pt_vulkan_api_test_bin': '//xplat'} To suppress this warning: touch ~/.config/.dont_hint_buck2 Building: finished in 0.1 sec (100%) 339/339 jobs, 0/339 updated Total time: 0.2 sec BUILD SUCCEEDED Running main() from third-party/googletest/1.14.0/googletest/googletest/src/gtest_main.cc Note: Google Test filter = *layer_norm* [==========] Running 10 tests from 1 test suite. [----------] Global test environment set-up. [----------] 10 tests from VulkanAPITest [ RUN ] VulkanAPITest.packed_layer_norm_2d [ OK ] VulkanAPITest.packed_layer_norm_2d (342 ms) [ RUN ] VulkanAPITest.packed_layer_norm_3d [ OK ] VulkanAPITest.packed_layer_norm_3d (284 ms) [ RUN ] VulkanAPITest.packed_layer_norm_4d [ OK ] VulkanAPITest.packed_layer_norm_4d (5 ms) [ RUN ] VulkanAPITest.layer_norm_invalid_inputs [ OK ] VulkanAPITest.layer_norm_invalid_inputs (28 ms) [ RUN ] VulkanAPITest.layer_norm_2d [ OK ] VulkanAPITest.layer_norm_2d (1 ms) [ RUN ] VulkanAPITest.layer_norm_3d [ OK ] VulkanAPITest.layer_norm_3d (2 ms) [ RUN ] VulkanAPITest.layer_norm_4d [ OK ] VulkanAPITest.layer_norm_4d (4 ms) [ RUN ] VulkanAPITest.native_layer_norm_2d [ OK ] VulkanAPITest.native_layer_norm_2d (1 ms) [ RUN ] VulkanAPITest.native_layer_norm_3d [ OK ] VulkanAPITest.native_layer_norm_3d (2 ms) [ RUN ] VulkanAPITest.native_layer_norm_4d [ OK ] VulkanAPITest.native_layer_norm_4d (6 ms) [----------] 10 tests from VulkanAPITest (679 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test suite ran. (679 ms total) [ PASSED ] 10 tests. ``` Full test result in P888496077, summary as below ``` [----------] 419 tests from VulkanAPITest (21652 ms total) [----------] Global test environment tear-down [==========] 419 tests from 1 test suite ran. (21652 ms total) [ PASSED ] 418 tests. [ SKIPPED ] 1 test, listed below: [ SKIPPED ] VulkanAPITest.querypool_flushed_shader_log ``` ## Graph representation comparison We created a model using `layer_norm` and traced it as below ``` class MyModel(torch.nn.Module): def __init__(self): super(MyModel, self).__init__() self.layer_norm = torch.nn.LayerNorm(normalized_shape=10) def forward(self, x): return self.layer_norm(x) # Create an instance of the model model = MyModel() # Create a dummy input tensor for tracing input_tensor = torch.randn(1, 10) # Use torch.jit.trace to trace the model and generate a graph traced_model = torch.jit.trace(model, input_tensor) ``` Then we converted the traced model to Vulkan backend using `optimize_for_mobile` ``` from torch.utils import mobile_optimizer vulkan_model = mobile_optimizer.optimize_for_mobile( traced_model, backend="vulkan", preserved_methods=to_preserve ) ``` Then we can print the graph of the `vulkan_model` as `print(vk_model.graph)` - Before this diff ``` %4 : bool = prim::Constant[value=1](), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0 %5 : float = prim::Constant[value=1.0000000000000001e-05](), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0 %14 : int[] = prim::Constant[value=[10]]() %33 : Tensor = aten::to(%x, %53, %30, %31, %31) %10 : Tensor = aten::layer_norm(%33, %14, %self.layer_norm.weight, %self.layer_norm.bias, %5, %4), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0 ``` - after this diff ``` %14 : int[] = prim::Constant[value=[10]]() %47 : Tensor = aten::to(%x, %78, %44, %45, %45) %16 : Tensor = vulkan_prepack::run_layernorm_context(%47, %14, %17) ``` Reviewed By: SS-JIA Differential Revision: D51530478

facebook-github-bot · 2023-11-29T17:03:36Z

This pull request was exported from Phabricator. Differential Revision: D51530478

facebook-github-bot · 2023-11-30T01:31:54Z

@pytorchbot merge

(Initiating merge automatically since Phabricator Diff has merged)

pytorchmergebot · 2023-11-30T01:33:40Z

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

Summary: `Layernorm` has two arguments weight and bias which are stored as constant tensors on the CPU and they are transferred to GPU at every inference call. We create a context for this op to avoid the repeated passing. Specifically, we - created `create_layernorm_context` and `run_layernorm_context` in `Layernorm.h` and `Layernorm.cpp` - registered them in `Register.cpp` - rewrote the graph representation of the op in `vulkan_rewrite.cpp` Test Plan: ## Numerical test ``` [luwei@devbig984.prn1 /data/users/luwei/fbsource (b6ccc956c)]$ LD_LIBRARY_PATH=third-party/swiftshader/lib/linux-x64/ buck run fbcode/mode/dev-nosan //xplat/caffe2:pt_vulkan_api_test_bin -- --gtest_filter="*layer_norm*" Recommended: For faster builds try buck2: replace 'buck' with 'buck2' NOTE: buck-out/ has changed: look for files in fbsource/buck-out/v2/ 'buck2 build --show-output //xplat/caffe2:pt_vulkan_api_test_bin' will print the new output paths. If you are building in fbsource//xplat and have questions, post in 'Cross Platform Dev Discussions': https://fb.workplace.com/groups/xplat.qa Targets matching .buckconfig buck2.supported_projects: {'//xplat/caffe2:pt_vulkan_api_test_bin': '//xplat'} To suppress this warning: touch ~/.config/.dont_hint_buck2 Building: finished in 0.1 sec (100%) 339/339 jobs, 0/339 updated Total time: 0.2 sec BUILD SUCCEEDED Running main() from third-party/googletest/1.14.0/googletest/googletest/src/gtest_main.cc Note: Google Test filter = *layer_norm* [==========] Running 10 tests from 1 test suite. [----------] Global test environment set-up. [----------] 10 tests from VulkanAPITest [ RUN ] VulkanAPITest.packed_layer_norm_2d [ OK ] VulkanAPITest.packed_layer_norm_2d (342 ms) [ RUN ] VulkanAPITest.packed_layer_norm_3d [ OK ] VulkanAPITest.packed_layer_norm_3d (284 ms) [ RUN ] VulkanAPITest.packed_layer_norm_4d [ OK ] VulkanAPITest.packed_layer_norm_4d (5 ms) [ RUN ] VulkanAPITest.layer_norm_invalid_inputs [ OK ] VulkanAPITest.layer_norm_invalid_inputs (28 ms) [ RUN ] VulkanAPITest.layer_norm_2d [ OK ] VulkanAPITest.layer_norm_2d (1 ms) [ RUN ] VulkanAPITest.layer_norm_3d [ OK ] VulkanAPITest.layer_norm_3d (2 ms) [ RUN ] VulkanAPITest.layer_norm_4d [ OK ] VulkanAPITest.layer_norm_4d (4 ms) [ RUN ] VulkanAPITest.native_layer_norm_2d [ OK ] VulkanAPITest.native_layer_norm_2d (1 ms) [ RUN ] VulkanAPITest.native_layer_norm_3d [ OK ] VulkanAPITest.native_layer_norm_3d (2 ms) [ RUN ] VulkanAPITest.native_layer_norm_4d [ OK ] VulkanAPITest.native_layer_norm_4d (6 ms) [----------] 10 tests from VulkanAPITest (679 ms total) [----------] Global test environment tear-down [==========] 10 tests from 1 test suite ran. (679 ms total) [ PASSED ] 10 tests. ``` Full test result in P888496077, summary as below ``` [----------] 419 tests from VulkanAPITest (21652 ms total) [----------] Global test environment tear-down [==========] 419 tests from 1 test suite ran. (21652 ms total) [ PASSED ] 418 tests. [ SKIPPED ] 1 test, listed below: [ SKIPPED ] VulkanAPITest.querypool_flushed_shader_log ``` ## Graph representation comparison We created a model using `layer_norm` and traced it as below ``` class MyModel(torch.nn.Module): def __init__(self): super(MyModel, self).__init__() self.layer_norm = torch.nn.LayerNorm(normalized_shape=10) def forward(self, x): return self.layer_norm(x) # Create an instance of the model model = MyModel() # Create a dummy input tensor for tracing input_tensor = torch.randn(1, 10) # Use torch.jit.trace to trace the model and generate a graph traced_model = torch.jit.trace(model, input_tensor) ``` Then we converted the traced model to Vulkan backend using `optimize_for_mobile` ``` from torch.utils import mobile_optimizer vulkan_model = mobile_optimizer.optimize_for_mobile( traced_model, backend="vulkan", preserved_methods=to_preserve ) ``` Then we can print the graph of the `vulkan_model` as `print(vk_model.graph)` - Before this diff ``` %4 : bool = prim::Constant[value=1](), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0 %5 : float = prim::Constant[value=1.0000000000000001e-05](), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0 %14 : int[] = prim::Constant[value=[10]]() %33 : Tensor = aten::to(%x, %53, %30, %31, %31) %10 : Tensor = aten::layer_norm(%33, %14, %self.layer_norm.weight, %self.layer_norm.bias, %5, %4), scope: __module.layer_norm # /mnt/xarfuse/uid-602118/33e18f68-seed-nspid4026531836_cgpid32066351-ns-4026531840/torch/nn/functional.py:2546:0 ``` - after this diff ``` %14 : int[] = prim::Constant[value=[10]]() %47 : Tensor = aten::to(%x, %78, %44, %45, %45) %16 : Tensor = vulkan_prepack::run_layernorm_context(%47, %14, %17) ``` Reviewed By: SS-JIA Differential Revision: D51530478 Pull Request resolved: pytorch#114701 Approved by: https://github.com/yipjustin

pytorch-bot bot added ciflow/periodic Trigger jobs ran periodically on master (periodic.yml) on the PR module: vulkan release notes: vulkan release notes category labels Nov 28, 2023

facebook-github-bot added the fb-exported label Nov 28, 2023

yipjustin approved these changes Nov 29, 2023

View reviewed changes

copyrightly force-pushed the export-D51530478 branch from aa81c68 to 6e7c759 Compare November 29, 2023 04:23

copyrightly force-pushed the export-D51530478 branch from 6e7c759 to f96df3b Compare November 29, 2023 17:03

pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label Nov 30, 2023

pytorchmergebot added the merging label Nov 30, 2023

pytorchmergebot added the Merged label Nov 30, 2023

pytorchmergebot closed this in 34ea0a2 Nov 30, 2023

pytorchmergebot removed the merging label Nov 30, 2023

facebook-github-bot deleted the export-D51530478 branch December 3, 2023 15:26

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[Pytoch][Vulkan] Create context for layernorm #114701

[Pytoch][Vulkan] Create context for layernorm #114701

Uh oh!

copyrightly commented Nov 28, 2023

Uh oh!

pytorch-bot bot commented Nov 28, 2023 •

edited

Loading

Uh oh!

facebook-github-bot commented Nov 28, 2023

Uh oh!

facebook-github-bot commented Nov 29, 2023

Uh oh!

facebook-github-bot commented Nov 29, 2023

Uh oh!

facebook-github-bot commented Nov 30, 2023

Uh oh!

pytorchmergebot commented Nov 30, 2023

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

[Pytoch][Vulkan] Create context for layernorm #114701

[Pytoch][Vulkan] Create context for layernorm #114701

Uh oh!

Conversation

copyrightly commented Nov 28, 2023

Numerical test

Graph representation comparison

Uh oh!

pytorch-bot bot commented Nov 28, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/114701

✅ No Failures

Uh oh!

facebook-github-bot commented Nov 28, 2023

Uh oh!

facebook-github-bot commented Nov 29, 2023

Uh oh!

facebook-github-bot commented Nov 29, 2023

Uh oh!

facebook-github-bot commented Nov 30, 2023

Uh oh!

pytorchmergebot commented Nov 30, 2023

Merge started

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

pytorch-bot bot commented Nov 28, 2023 •

edited

Loading