Abstract common helpers from CUDA implementation for code reusing in upcoming SYCL kernels #117234
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/117234
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit d0f7eaa with merge base b830751 ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
fengyuan14
added a commit
that referenced
this pull request
Jan 11, 2024
…upcoming SYCL kernels e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> ghstack-source-id: 0db78a6 Pull Request resolved: #117234
…upcoming SYCL kernels e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <[email protected]> [ghstack-poisoned]
jgong5
approved these changes
Jan 12, 2024
fengyuan14
added a commit
that referenced
this pull request
Jan 12, 2024
…upcoming SYCL kernels e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> ghstack-source-id: 9cafc7c Pull Request resolved: #117234
…reusing in upcoming SYCL kernels" e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> [ghstack-poisoned]
This was referenced Jan 23, 2024
Closed
…reusing in upcoming SYCL kernels" e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> [ghstack-poisoned]
…reusing in upcoming SYCL kernels" e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> [ghstack-poisoned]
…reusing in upcoming SYCL kernels" e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> [ghstack-poisoned]
…reusing in upcoming SYCL kernels" e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> [ghstack-poisoned]
fengyuan14
added a commit
that referenced
this pull request
Jan 24, 2024
…upcoming SYCL kernels e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> ghstack-source-id: 9634e48 Pull Request resolved: #117234
…reusing in upcoming SYCL kernels" e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> [ghstack-poisoned]
…reusing in upcoming SYCL kernels" e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> [ghstack-poisoned]
gujinghui
approved these changes
Jan 31, 2024
Collaborator
|
@pytorchbot rebase |
Collaborator
|
@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here |
…reusing in upcoming SYCL kernels" e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> [ghstack-poisoned]
Collaborator
|
Successfully rebased |
pytorchmergebot
pushed a commit
that referenced
this pull request
Jan 31, 2024
…upcoming SYCL kernels e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> ghstack-source-id: e619ad4 Pull Request resolved: #117234
fengyuan14
added a commit
that referenced
this pull request
Feb 5, 2024
…upcoming SYCL kernels e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> ghstack-source-id: 8943663 Pull Request resolved: #117234
…reusing in upcoming SYCL kernels" e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke. feature: #114835 Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke. They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for, when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency. They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2) and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size). Signed-off-by: Feng Yuan <feng1.yuanintel.com> [ghstack-poisoned]
Contributor
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
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.
Stack from ghstack (oldest at bottom):
e.g. OffsetCalculator, IntegerDivider, MemoryAccess utilities, ElementwiseInvoke.
feature: #114835
Abstracted helpers are about integer arithmetic and C++ standard template helpers for load, store and custom functor invoke.
They are general for any kind of kernel language, which supports basic interger arithmetic and C++ standard template, except for,
when we have some backend specific intrinsic, like CUDA __umulhi. Using backend specific macro to isolate divergency.
They are irrelevant to concurrency algorithm of kernel, backend specific data type (e.g. CUDA half2)
and backend specific logic concurrent resource configuration (e.g. CUDA block size, thread work size).
Signed-off-by: Feng Yuan [email protected]