Enable Dynamic shape support via tensor virtual and physical resizing #2340
Conversation
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/2340
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New FailuresAs of commit 7be9040 with merge base e5ef31d ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Mar 10, 2024
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. Differential Revision: D54728401
SS-JIA
force-pushed
the
export-D54728401
branch
from
e820fee to
c3990de
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Mar 10, 2024
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. Differential Revision: D54728401
SS-JIA
force-pushed
the
export-D54728401
branch
from
c3990de to
c98c726
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Mar 10, 2024
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. Differential Revision: D54728401
SS-JIA
force-pushed
the
export-D54728401
branch
from
c98c726 to
b0d5a6f
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Mar 10, 2024
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. Differential Revision: D54728401
SS-JIA
force-pushed
the
export-D54728401
branch
from
b0d5a6f to
461bc8f
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Mar 11, 2024
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. Differential Revision: D54728401
SS-JIA
force-pushed
the
export-D54728401
branch
from
461bc8f to
375480b
Compare
jorgep31415
approved these changes
Mar 11, 2024
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Mar 11, 2024
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. Reviewed By: jorgep31415 Differential Revision: D54728401
SS-JIA
force-pushed
the
export-D54728401
branch
from
375480b to
758449b
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
jorgep31415
added a commit
that referenced
this pull request
Mar 11, 2024
Before: Each node contains a `UniformParamsBuffer`. After: Each node contains a `std::vector<std::shared_ptr<UniformParamsBuffer>>`. In follow up changes, we will break up parameters to be passed via multiple UniformParamsBuffer, since 1. some are tensor-specific (e.g. image extents) and 2. others are operator-specific (e.g. alpha for binary ops). Hence, we need **`std::vector`**. We are adding the methods for #1 in #2340. Since #1 and #2 will be owned by different objects, we need **pointers**. Since #1 is owned by `vTensor` which is non-copyable, we can't use unique_ptr so we need **`std::shared_ptr`**. Differential Revision: [D54691831](https://our.internmc.facebook.com/intern/diff/D54691831/) [ghstack-poisoned]
jorgep31415
added a commit
that referenced
this pull request
Mar 11, 2024
Before: Each node contains a `UniformParamsBuffer`. After: Each node contains a `std::vector<std::shared_ptr<UniformParamsBuffer>>`. In follow up changes, we will break up parameters to be passed via multiple UniformParamsBuffer, since 1. some are tensor-specific (e.g. image extents) and 2. others are operator-specific (e.g. alpha for binary ops). Hence, we need **`std::vector`**. We are adding the methods for #1 in #2340. Since #1 and #2 will be owned by different objects, we need **pointers**. Since #1 is owned by `vTensor` which is non-copyable, we can't use unique_ptr so we need **`std::shared_ptr`**. Differential Revision: [D54691831](https://our.internmc.facebook.com/intern/diff/D54691831/) ghstack-source-id: 218195447 Pull Request resolved: #2348
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Mar 11, 2024
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. bypass-github-export-checks Reviewed By: jorgep31415 Differential Revision: D54728401
SS-JIA
force-pushed
the
export-D54728401
branch
from
758449b to
a6a726e
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
SS-JIA
added a commit
to SS-JIA/pytorch
that referenced
this pull request
Mar 11, 2024
…resizing (pytorch#121598) Summary: X-link: pytorch/executorch#2340 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. bypass-github-export-checks Test Plan: Check CI. Tests have also been added to `vulkan_compute_api_test` that test the two methods of tensor resizing. Reviewed By: jorgep31415 Differential Revision: D54728401
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Mar 11, 2024
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. bypass-github-export-checks Reviewed By: jorgep31415 Differential Revision: D54728401
SS-JIA
force-pushed
the
export-D54728401
branch
from
a6a726e to
ad3458c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
facebook-github-bot
pushed a commit
that referenced
this pull request
Mar 11, 2024
Summary: bypass-github-export-checks Pull Request resolved: #2348 Before: Each node contains a `UniformParamsBuffer`. After: Each node contains a `std::vector<std::shared_ptr<UniformParamsBuffer>>`. In follow up changes, we will break up parameters to be passed via multiple UniformParamsBuffer, since 1. some are tensor-specific (e.g. image extents) and 2. others are operator-specific (e.g. alpha for binary ops). Hence, we need **`std::vector`**. We are adding the methods for #1 in #2340. Since #1 and #2 will be owned by different objects, we need **pointers**. Since #1 is owned by `vTensor` which is non-copyable, we can't use unique_ptr so we need **`std::shared_ptr`**. ghstack-source-id: 218195447 exported-using-ghexport Reviewed By: SS-JIA Differential Revision: D54691831 fbshipit-source-id: 84ab9f777e247fd56234290ed7f7343b9701c73f
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Mar 12, 2024
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. bypass-github-pytorch-ci-checks bypass-github-executorch-ci-checks bypass-github-export-checks Reviewed By: jorgep31415 Differential Revision: D54728401
SS-JIA
force-pushed
the
export-D54728401
branch
from
ad3458c to
ff1fa4f
Compare
…pytorch#2340) Summary: X-link: pytorch/pytorch#121598 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. bypass-github-pytorch-ci-checks bypass-github-executorch-ci-checks bypass-github-export-checks Reviewed By: jorgep31415 Differential Revision: D54728401
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
SS-JIA
force-pushed
the
export-D54728401
branch
from
ff1fa4f to
7be9040
Compare
|
This pull request was exported from Phabricator. Differential Revision: D54728401 |
SS-JIA
added a commit
to SS-JIA/pytorch
that referenced
this pull request
Mar 12, 2024
…resizing (pytorch#121598) Summary: X-link: pytorch/executorch#2340 ## Context This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways: 1. Discarding underlying `vkImage` or `vkBuffer` and reallocating a new `vkImage` or `vkBuffer` with updated sizes. This method is intended to be used when the current `vkImage` or `vkBuffer` is not large enough to contain the new sizes. 2. Update the tensor's size metadata without reallocating any new resources. This allows shaders to interpret the underlying `vkImage` or `vkBuffer` as if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed. bypass-github-pytorch-ci-checks bypass-github-executorch-ci-checks bypass-github-export-checks Test Plan: Check CI. Tests have also been added to `vulkan_compute_api_test` that test the two methods of tensor resizing. Reviewed By: jorgep31415 Differential Revision: D54728401
|
This pull request has been merged in 4dfb637. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Summary:
Context
This changeset lays the foundations for supporting dynamic shapes in the ExecuTorch Vulkan delegate via allowing Tensors to be resized in one of two ways:
vkImageorvkBufferand reallocating a newvkImageorvkBufferwith updated sizes. This method is intended to be used when the currentvkImageorvkBufferis not large enough to contain the new sizes.vkImageorvkBufferas if it were smaller than it actually is, and allows command buffers to be preserved when sizes are changed.Differential Revision: D54728401